Skip to main content

Full text of "The Moral Judgment Of The Child"

See other formats

136*72 P58te 6125348 

The moral of the child 



change of residence pm 

:: ^v : "i^^-W 


3 1148 0045' 


' QCT 6 J19f7 
OCT 6 ^i' 

The Moral Judgment 
of the Child 

The Moral Judgment 
of the Child 

B 7 


Doctor of Science > Professor at ike University of Geneva, Director of the 

International Bureau of Education, Co-Director of the Institut 

f. J. Rousseau^ Geneva; Author of "Language and Thought 

of the Child t " "Judgment and Reasoning in the Child" 

"The Child's Conception of the World," "The 

Child** Conception of Causality " 



Translated by 

Printed in U.S.A, 



FOREWORD ........ vii 


i. The rules of the game of marbles, p. 4. 2. The 
interrogatory and its general results, p. 13. 3. The 
practice of rules. I. The first two stages, p. 19. 4. The 
practice of rules. II. Third and fourth stages, p. 32. 
5. Consciousness of rules. I. The first two stages, 
p. 41. 6. The consciousness of rules. II. Third stage, 
p. 56. 7. A girls' game : " flet cachant ", p. 69. 
8. Conclusion : I. Motor rules and the two kinds of 
respect, p. 76. 9. Conclusion : II. Respect for the 
group or respect for persons. Search for a guiding 
hypothesis, p. 95. 


REALISM ..... 104 

i. The method, p. 107. 2. Objective responsibility. 

I. Clumsiness and stealing, p. 116. 3. Objective 
responsibility. II. Lying, p. 135. 4. Lying and the 
two kinds of respect, p. 159. 5. Conclusion. Moral 
realism, p. 171. General conclusion, p. 192. 


i. The problem of punishments and retributive justice, 
p. 197. 2. Collective and communicable responsibility, 
p. 231. 3. Immanent Justice, p. 250. 4. Retributive 
justice and distributive justice, p. 262.- 5. Equality 
and authority, p, 275. 6. Justice between children, 
p. 297. 7, Conclusion ; the idea of justice, p. 312. 



i. The theories of Durkheim and Fauconnet on respon- 
sibility, p. 327. 2. Durkheim 's doctrine of moral 
authority. I. Introduction. Durkheim's, p. 341.- 
3. The theory of authority according to Durkheim. 

II. Moral education, p. 355. 4- M. Pierre liovet's 
theory, p. 375. - 5, The point of view of J, M. Baldwin, 
p. 392.- 6. General conclusions, p. 401. 


INDEX OF NAMES ...... 418 

CM 2 f> 3 4 8 


READERS will find in this book no direct analysis of child 
morality as it is practised in home and school life or in 
children's societies. It is the moral judgment that we 
propose to investigate, not moral behaviour or sentiments. 
With this aim in view we questioned a large number of 
children from the Geneva and Neuchcltel schools and 
held conversations with them, similar to those we had 
had before on their conception of the world and of 
causality. The present volume contains the results of 
these conversations. 

First we had to establish what was meant by respect 
for rules from the child's point of view. This is why we 
have begun with an analysis of the rules of a social game 
in the obligatory aspect which these possess for a bona 
fide player. From the rules of games we have passed 
to the specifically moral rules laid down by adults and 
we have tried to see what idea the child forms of these 
particular duties. Children's ideas on lying were selected 
as being a privileged example. Finally we have examined 
the notions that arose out of the relations in which the 
children stood to each other and we were thus led to 
discuss the idea of justice as our special theme. 

Having reached this point, our results seemed to us 
sufficiently consistent to be compared to some of the 
hypotheses now in favour among sociologists and writers 
on the psychology of morals. It is to this final task 
that we have devoted our fourth chapter. 

We are more conscious than anybody of the defects 
as of the advantages of the method we have used. The 
great danger, especially in matters of morality, is that 
of making the child say whatever one wants him to say. 




There is no infallible remedy for this ; neither the good 
faith of the questioner nor the precautionary methods 
which we have laid stress upon elsewhere * are sufficient. 
The only safeguard lies in the collaboration of other 
investigators, If other psychologists take up our questions 
from different view-points and put them to children of 
differing social environment, it will be possible sooner or 
later to separate the objective from the arbitrary elements 
in the results which we bring forward in this work. An 
analogous task has been undertaken in various countries 
with regard to child logic and children's ideas on causality ; 
and while certain exaggerations of which we had been 
guilty came to light in this way, the results up to date 
in no way tend to discourage us in the use of the method 
we have adopted. 

The advantages of this method seem to us to be that 
it makes evident what observation left to itself can only 
surmise. During the last few years, for example, I have 
been engaged in taking down the spontaneous remarks 
made by my two little girls, to whom 1 have never set 
the questions examined in The Child's Conception of the 
World or in The Child's Conception of Causality. Now, 
"broadly speaking, the tendencies to Realism, Animism, 
Artificialism and dynamic Causality, etc., come very 
clearly to light, but the meaning of these children's most 
interesting " whys ", as of many of their chance remarks, 
would have almost completely eluded me if I had not in 
the past questioned hundreds of children personally on 
the same subjects. A child's spontaneous remark is, of 
course, more valuable than all the questioning in the 
world. But in child psychology such a remark cannot 
be seen in its right perspective without the work of 
preparation constituted by those very interrogatories. 

i See The Child's Conception of the World, Kegan Paul, which in the 
sequel will be designated by the letters C.W. My other books, Language 
and Thought in the Child, Judgment and Reasoning in the Child, and 
The Child's Conception of the World, will be referred to by the initials 
L.T., /.JR., and C.VK. respectively. 


The present book on child morality is just such a 
preliminary piece of work. It is my sincere hope tljat 
it may supply a scaffolding which those living with 
children and observing their spontaneous reactions can 
use in erecting the actual edifice. In a sense, child 
morality throws light on adult morality. If we want to 
form men and women nothing will fit us so well for the 
task as to study the laws that govern their formation. 

The Moral Judgment of the Child 


CHILDREN'S games constitute the most admirable social 
institutions. The game of marbles, for instance, as 
played by boys, contains an extremely complex system 
of rules, that is to say, a code of laws, a jurisprudence of 
its own. Only the psychologist, whose profession obliges 
him to become familiar with this instance of common 
law, and to get at the implicit morality underlying it, is 
in a position to estimate the extraordinary wealth of these 
rules by the difficulty he experiences in mastering their 

If we wish to gain any understanding of child morality, 
it is obviously with the analysis of such facts as these 
that we must begin. All morality consists in a system 
of rules, and the essence of all morality is to be sought 
for in the respect which the individual acquires for these 
rules. The reflective analysis of Kant, the sociology of 
Durkheim, or the individualistic psychology of Bovet 
all meet on this point. The doctrines begin to diverge 
only from the moment that it has to be explained how 
the mind comes to respect these rules. For our part, 
it will be in the domain of child psychology that we 
shall undertake the analysis of this " how ". 

Now, most of the moral rules which the child learns 
to respect he receives from adults, which means that he 

i With the collaboration of Mme V. J, Piaget, MM M. Lambercier 
and L. Martinez. 

A I 


receives them after they have been fully elaborated, and 
often elaborated, not in relation to him and as they are 
needed, but once and for all and through an uninterrupted 
succession of earlier adult generations. 

In the case of the very simplest social games, on the 
contrary, we are in the presence of rules which have been 
elaborated by the children alone. It is of no moment 
whether these games strike us as " moral " or not in 
their contents. As psychologists we must ourselves adopt 
the point of view, not of the adult conscience, but of 
child morality. Now, the rules of the game of marbles 
are handed down, just like so-called moral realities, from 
one generation to another, and are preserved solely by the 
respect that is -felt for them by individuals. The sole 
difference is that the relations in this case are only those 
that exist between children. The little boys who are 
beginning to play are gradually trained by the older ones 
in respect for the law ; and in any case they aspire from 
their hearts to the virtue, supremely characteristic of 
human dignity, which consists in making a correct use of 
the customary practices of a game. As to the older 
ones, it is in their power to alter the rules. If this is not 
" morality ", then where does morality begin ? At least, 
it is respect for rules, and it appertains to an enquiry 
like ours to begin with the study of facts of this order. 
Of course the phenomena relating to the game of marbles 
are not among the most primitive. Before playing with 
his equals, the child is influenced by his parents. He is 
subjected from his cradle to a multiplicity of regulations, 
and even before language he becomes conscious of certain 
obligations. These circumstances even exercise, as we 
shall see, an undeniable influence upon the way in which 
the rules of games are elaborated. But in the case of 
play institutions, adult intervention is at any rate reduced 
to the minimum. We are therefore in the presence here 
of realities which, if not amongst the most elementary, 
should be classed nevertheless amongst the most spon- 
taneous and the most instructive. 


With regard to game rules there are two phenomena 
which it is particularly easy to study : first the practice 
of rules, i.e. the way in which children of different ages 
effectively apply rules : second the consciousness of rules, 
i.e. the idea which children of different ages form of 
the character of these game rules, whether of something 
obligatory and sacred or of something subject to their 
own choice, whether of heteronomy or autonomy. 

It is the comparison of these two groups of data which 
constitutes the real aim of this chapter. For the relations 
which exist between the practice and the consciousness 
of rules are those which will best enable us to define the 
psychological nature of moral realities. 

One word more. Before embarking upon an analysis 
of the practice or of the consciousness of rules, we must 
first give some account of the actual content of these 
rules. We must therefore establish the social data of the 
problem. But we shall confine ourselves only to what 
is indispensable. We have not attempted to establish 
the sociology of the game of marbles ; this would 
have meant finding out how this game was played in 
the past and how it is now played in different parts of 
the world (it is actually played by negro children). Even 
confining ourselves to French Switzerland, we believe 
it would need several years of research to discover all 
the local variants of the game and, above all, to outline 
the history of these variants throughout the last few 
generations. Such an enquiry, which might be useful to 
the sociologist, is superfluous for the psychologist. All 
the latter needs in order to study how rules are learned 
is a thorough knowledge of a given custom in actual 
use, just as in order to study child language, all he needs 
is to know a given dialect, however localized, without 
troubling to reconstruct all its semantic and phonetic 
changes in time and space. We shall therefore confine 
ourselves to a short analysis of the content of the game 
as it is played in Geneva and NeucMtel, in the districts 
where we conducted our work. 


essential facts must be noted if we wish to analyse 
simultaneously the practice and the consciousness of rules. 

The first is that among children of a given generation 
and in a given locality, however small, there is never one 
single way of playing marbles, there are quantities of 
ways. There is the " square game " with which we shall 
occupy ourselves more especially. A square is drawn on 
the ground and a number of marbles placed within it ; 
the game consists in aiming at these from a distance and 
driving them out of the enclosure. There is the game of 
" courate " where two players aim at each other's marble 
in indefinite pursuit. There is the game of " troyat " 
from " trou " (=hole) or " creux " (= hollow), where the 
marbles are piled into a hole and have to be dislodged 
by means of a heavier marble, and so on. Every child 
is familiar with several games, a fact that may help 
according to his age to reinforce or to weaken his belief 
in the sacred character of rules. 

In the second place, one and the same game, such as 
the Square game, admits of fairly important variations 
according to when and where it is played. As we had 
occasion to verify, the rules of the Square game are not 
the same in four of the communes of NeucMtel l situated 
at 2-3 kilometres from each other. They are not the 
same in Geneva and in Neuchtel. They differ, on 
certain points, from one district to another, from one 
school to another in the same town. In addition to this, 
as through our collaborators 1 kindness we were able to 
establish, variations occur from one generation to another. 
A student of twenty assured us that in his village the 
game is no longer played as it was " in his days ". These 
variations according to time and place are important, 
because children are often aware of their existence. A 
child who has moved from one town, or merely from one 
school building to another will often explain to, us that 
such and such a rule is in force in one place but not in 

1 Neuch&tel, La Coudre, Hauterive and Saint-Blaise. 


tlie other. Very often, too, a child will tell us that his 
father played differently from him. Last of all, there 
is the boy of 14 who has given up playing because he is 
beginning to feel superior to the little ones, and who, 
according to his temperament,. laughs or mouras over the 
fact that the customs of his generation are going by the 
board instead of being piously preserved by the rising 

Finally, and clearly as a result of the convergence of 
these local or historical currents, it will happen that one 
and the same game (like the Square game) played in the 
playground of one and the same school admits on certain 
points of several different rules. Children of n to 13 are 
familiar with these variants, and they generally agree 
before or during the game to choose a given usage to the 
exclusion of others. These facts must therefore be borne 
in mind, for they undoubtedly condition the judgment 
which the child will make on the value of rules. 

Having mentioned these points, we shall give a brief 
exposition of the rules of the Square game, which will 
serve as a prototype, and we shall begin by fixing the 
child's language so as to be able to understand the reports 
of the conversations which will be quoted later on. 
Besides, as is so often the case in child psychology, some 
aspects of this language are in themselves highly in- 

A marble is called " un marbre " in NeucMtel and " un 
coeillu " or " un mapis " in Geneva. There are marbles 
of different value. The cement marble has the place of 
honour. The " carron " which is smaller and made of 
the more brittle clay is of less value because it costs less. 
The marbles that are used for throwing l and are not 
placed inside the square are called according to their 
consistency " corna " (if in carnelian), " ago ", or " agathe ", 

1 The English technical equivalent is the generic term " shooter " 
which we shall use in the interrogatories given below. For the rest 
we have generally retained the French words as one cannot be sure 
that the English terms mean exactly the same. [Trans.] 


" cassine " (glass ball with coloured veins), " plomb " 
(large marble containiBg lead), etc. Each is worth so 
many marbles or so many " canons ". To throw a 
marble is to " tirer " (shoot) and to touch another marble 
with ones own is to " tanner " (hit). 

Then coifies a set of terms of ritual consecration, that is, 
of expressions which the player uses in order to announce 
that he is going to perform such-and-such an operation and 
which thus consecrate it ritually as an accomplished fact. 
For, once these words have been uttered, the opponent 
is powerless against his partner's decision ; whereas if he 
takes the initiative by means of the terms of ritual 
interdiction, which we shall examine in a moment, he will 
in this way prevent the operation which he fears. For 
example, in order to play first in circumstances when it 
is possible to do so, the child will say (at NeucMtel) 
lt prems " obviously a corruption of the word ' ' premier " 
(first). If he wants to go back to the line that all the 
players start from at fheir first turn and which is called 
the " coche 'V he simply says " coche ". If he wishes 
to advance or retreat to a distance twice as great, he 
says " deux coches ", or if to a distance of one, two, or 
three hand-breadths he says " one (or two, or three) 
empans " (spans). If he wishes to place himself in 
relation to the square at a distance equal to that at 
which he finds himself at a given moment, but in another 
direction (so as to avoid the probable attacks of his 
opponent) he says " du mien" (mine), and if he wishes 
to prevent Ms opponent from doing the same thing he 
says " du tien " (yours). This applies to Neuchatel. 
In Geneva these displacements are expressed by the 
terms " faire une entasse " or " entorse " (to make a 
twist). If you wish to give up your turn and be " dead " 
until your opponent has moved, you say '" coup passe " 
(my turn passed). 

As soon as these terms have been uttered in circum- 
stances which of course are carefully regulated by a 

1 English, pitch-line (sometimes). [Trans.] 


whole juridical system, the opponent has to submit. 
But if the opponent wishes to anticipate these operations, 
it is sufficient for him to pronounce the terms of ritual 
interdiction, which at Neuchatel are simply the same 
terms but preceded by the prefix " fan ", from " defendu " 
(forbidden). For example, " fan-du-mien **, " fan-du- 
tien ", " fan-coche ", " fan-coup-passe ", etc. Some 
children, not having understood this prefix, which does 
not, after all, correspond with anything in the speech they 
hear around them, say " femme-du-tien ", " femme- 
coche ", etc. 

Two more particularly suggestive terms of consecration 
should be noted, which are .current among the little 
Genevans : " glaine " and " toumike ". When a player 
places a marble of superior value in the square, thinking 
that he has put down an ordinary marble (say an " ago " 
instead of a " coeillu ") he is naturally allowed, if he has 
noticed his mistake, to pick up his " ago " and put an 
ordinary marble in its place. Only a dishonest opponent 
would take advantage of his partner's absent-mindedness 
and pocket this " ago " after having hit it. The children 
we questioned on this point were unanimous in pronounc- 
ing such procedure equivalent to stealing. But if, on the 
other hand, the opponent spots his partner's mistake in 
time and utters the word " toumike " or (by doubling 
the last syllable) " toumikemik ", then the absent-minded 
player no longer has the right to pick up his " ago " ; 
he must leave it on the ground like a common-or-garden 
" coeillu ", and if one of the players succeeds in hitting it, 
this player will be allowed in all fairness to take possession 
of it. This shows us a very interesting example of a 
word consecrating a mistake and by doing so changing 
a dishonest action into one that is legitimate and recognized 
as such by all. We have here for the first time an example 
of that formalism, which belongs to certain aspects of 
childish morality, and into whose nature we shall go more 
deeply in the sequel in connection with objective 


In the same way, the word "glaine" legitimatizes 
piracy in certain well-defined conditions. When one of 
the players has succeeded, either by luck or by skill, 
in winning all his partners' marbles, it is a point of honour 
similar to that which sociologists designate with the 
term " potlatch " that he should offer to play a fresh 
set and should himself place in the square the necessary 
marbles, so as to give his less fortunate playmates the 
chance of recovering a portion of their possessions. If 
he refuses, of course no law can force him to do this ; 
he has won and there is the end of it. If, however, one 
of the players pronounces the word " glaine " then the 
whole gang falls- upon the miser, throws him down, 
empties his pockets and shares the booty. This act of 
piracy which in normal times is profoundly contrary to 
morality (since the marbles collected by the winner 
constitute his lawfully acquired possession) is thus changed 
into a legitimate act and even into an act of retributive 
justice approved by the general conscience when the 
word " glaine " has been pronounced. 1 

At NeucMtel we noticed neither " glaine " nor " tou- 
mik ", but, on the other hand, we found " cougac ". 
When one of the players has won too much (therefore in 
the situation just described) his defeated partner can 
force him to offer to play another set by uttering the 
word " cougac " (probably derived from coup-gagne just 
as " prems " was from premier). If the winner wishes to 
evade the obligation laid upon him by the fateful word, 
he has only to anticipate the blow by saying " fan-cougac ". 

Our reason for emphasizing these linguistic peculiarities 
is only to show from the first the juridical complexity of 
game rules. It is obvious that these facts could be 
analysed more fundamentally from other points of view. 

1 This word, " glaine " really has a wider sense. According to several 
children it entitles whoever pronounces it simply to pick up all the 
marbles that are on the ground when a discussion arises about them, 
or if a player forgets to take possession of what is his due. It is in 
this sense that the word is taken, for instance, in Philippe Monnier's, 
L* Livre de Blaise (3rd ed., p. 135). 


One could, for example, work out the whole psychology 
of consecration and interdiction in connection with the 
child and, above all, the psychology of social games. 
But these questions are really outside our scope. 1 Let us 
therefore return to what is the essential point so far as we 
are concerned, namely, the rules themselves. 

The Square game thus consists, in a word, in putting 
a few marbles in a square, and in taking possession of 
them by dislodging them with a special marble, bigger 
than the rest. But when it comes to details this simple 
schema contains an indefinite series of complications. Let 
us take them in order, so as to get some idea of their 

First of all, there is the " pose J> or outlay. One 
of the players draws a square and then each places his 
" pose ". If there are two players, each one puts down 
two, three, or four marbles. If there are three players, 
each puts down two marbles. If there are four or more 
players, it is customary to put down only one marble 
each. The main thing is equality : each one puts down 
what the others do. But in order to reach equality the 
relative value of the marbles must be taken into account. 
For an ordinary marble, you must put down eight 
" carrons ". A little " corna " is worth eight " marbres ", 
sixteen " carrons ", and so on. The values are carefully 
regulated and correspond roughly to the price paid at 
the shop round the corner. But alongside of financial 
operations proper, there are between children various 
exchanges in kind which appreciably alter current 

Then the game begins. A certain distance is agreed 
upon where the " coche " is drawn ; this is the line from 
which the players start. It is drawn parallel to and generally 
one or two metres away from one of the sides of the 
square, and from it each player will fire his first shot. 

1 With regard to social games we are awaiting the publication of 
R. Cousinet's book which will incorporate all the valuable material which 
this author has been accumulating for, so many years. 


(To " fire " is to throw one's shooter" agathe " or 
" comaline " into the square.) 

All, therefore, start from the coche. In some games 
yon return to the coche at each fresh turn, but it is more 
usual after the first shot to play from the place that your 
marble has rolled to. Sometimes this rule is limited by 
saying that the marble must not be further removed 
from the square than the coche. Thus if your marble 
has rolled two metres away from the square in any 
direction whatsoever, you bring- it back to a distance of 
im. 50 if this is the distance at which the coche itself 

But before the game begins you must settle who is to 
play first. For the first player has the advantage of 
" firing " into a square full of marbles, whereas ,'those 
who follow are faced only with what is left after the 
gains of the preceding players. In order to know who 
is to begin, a series of well-known rites are put in action. 
Two children walk towards each other stepping heel to 
toe, and whichever steps on the other's toe has the right 
to begin. Or else rhymed formula or even syllables 
devoid of any meaning are recited in sacramental order. 
Each syllable corresponds to a player, and he on whom 
the last syllable falls is the lucky one. In addition to 
these customary usages there is a method of procedure 
peculiar to the game of marbles. Each boy throws his 
" shooter " in the direction of the coche or of a line 
specially traced for the purpose. Whoever comes nearest 
up to the line begins. The others follow in order of 
their nearness up to the line. The last to play is the 
boy who has gone beyond the coche, and if several have 
gone beyond it, the last to play will be the boy whose 
marble has gone furthest. 

The order of the players having been settled in this 
way, the game begins. Each player in turn stands behind 
the coche and " fires " into the square. There are three 
ways of throwing one's marble : " Piquette " (Engl., 
" shooting ") which consists in projecting the marble by 


a jerk of the thumb, the marble being placed against the 
thumb-nail and kept in place by the first finger ; " Rou- 
lette " (Engl., " bowling J> ) which consists simply in 
rolling your marble along the ground, and " Poussette " 
(EngL, " hunching ") which consists in addition in carrying 
your hand along with it over a sufficient distance to correct 
the initial direction. Poussette is always banned and 
may in this connection be compared to the push stroke 
of a bad billiard player. At Neuchatel it is customary 
to say " fan-poussette " or again " femme-poussette ". 
In Geneva, the simpler expression " defendu de trainer " 
(dragging forbidden) is in use. Roulette (" bowling ") 
is also generally banned (" fan-roulette ") but is at times 
tolerated, in which case everyone will of course have the 
right to play in this way, and absolute equality before 
the law will even be agreed upon at the beginning of the 

The players are therefore throwing in the manner that 
has been agreed upon. Suppose one of the marbles 
included in the square has been hit. If it has gone outside 
the square it becomes the property of the boy who has 
dislodged it. If it remains inside the enclosure it cannot 
be taken. If, finally, it remains on the line the case is 
judged by the partners : a marble which is half outside 
is regarded as out, not otherwise. Here, naturally, a 
whole lot of subsidiary rules will establish the procedure 
in disputed cases. There remains the case of the marble 
with which one shoots (the shooter, or taw, etc.) remaining 
in the square or failing to lie beyond one of the lines of 
the square by at least half of its diameter : its owner 
is " cuit " (dished), i.e. he cannot play any more. If this 
marble is projected outside the square by that of another 
player, it becomes, like the others, the latter's property, 
except in the case of special conventions generally agreed 
upon at the beginning of the game. Finally, there are 
the possible complications arising from cases of rebounding 
marbles. A marble that bounces out of the square off 
another is sometimes not held to be won, and a fortiori 


in the case of a marble of value. 1 In other cases, every- 
'thing that goes outside the enclosure belongs to the 
player who has expelled it. The particular cases that 
arise in this way are settled in conformity with principles 
that are established either before or during the game by 
mutual agreement between all the participants. 

Then comes the question of the number of " shots " 
to be allowed to each. The player who has succeeded 
in winning one or more marbles has the right to play 
again, and so on, for as long as he wins. But sometimes 
the following reservation is made : for the first round in 
each game every player plays once in turn, independently 
of gains or losses. Here again, therefore, it is a matter 
of previous arrangement. 

In addition and this is an essential rule everyone 
has the right not only to " fire " at the marbles in the 
square; but also to " tanner " (hit) his neighbour's shooter, 
even outside the enclosure and indeed wherever it may 
happen to be in the course of the game. And of course 
the great difficulty is to shoot at the square without 
placing yourself within reach of your partners. This is 
why, when a shot would involve too many risks, you are 
allowed to say " coup-passe " and to remain where you 
are, provided, of course, that no one has foreseen this 
decision and said " fan-coup-passe ". And this, really, 
is why you are allowed to change your position provided 
you place yourself at the same distance from the square 
as before, and provided you first say " du mien " (mine), 
unless, once again, your opponent has anticipated your 
move by saying " du tien " (yours). 

Finally, a series of special rules deserves mention, the 
observance of which depends upon the particular town or 
school in question. The first player who says " place- 
pour-moi " (place for me) is not obliged to take up his 
position at one of the corners of the square. Any player 
who has succeeded in winning the equivalent of his 
" pose " (i.e. two marbles if he has placed two in the 

1 This is expressed by saying that the "revenette " does not count. 


square, and so on) can say " queue-de-pose " which will 
allow him. to have the first shot from the coche in the 
next game, and so on. 

The game, regulated in this way by an indefinite 
number of rules, is carried on until the square is empty. 
The boy who has pocketed the largest number of marbles 
has won, 


The rules that we have outlined above constitute a well- 
marked social reality, " independent of individuals " (in 
Durkheim's sense) and transmitted, like a language, from 
one generation to another. This set of customs is obviously 
more or less plastic. But individual innovations, just as 
in the case of language, succeed only when they meet 
a general need and when they are collectively sanctioned 
as being in conformity with the " spirit of the game ". 
But while fully recognizing the interest attaching to 
this sociological aspect of the problem, it was from a 
different standpoint that we raised the questions which 
we are now going to study. We simply asked ourselves 
(i) how the individuals adapt themselves to these rules, 
i.e. how they observe rules at each age and level of 
mental development ; (2) how far they become conscious 
of rules, in other words, what types of obligation result 
(always according to the children's ages) from the increasing 
ascendancy exercised by rules. 

The interrogatory is therefore easy to carry out. During 
the first part, it is sufficient to ask the children (we 
questioned about 20 boys ranging from 4 to 12-13) how 
one plays marbles. The experimenter speaks more or 
less as follows. " Here-are some marbles ". (The marbles 
are placed on a large baize-covered table beside a piece 
of chalk.) " You must show me how to play. When I 
was little I used to play a lot, but now I've quite for- 
gotten how to. I'd like to play again. Let's play together. 
You'll teach me the rules and I'll play with you." The 
child then draws a square, takes half the marbles, puts 


down his " pose ", and the game begins. It is important 
to bear in mind all possible contingencies of the game and 
to ask the child about each. This means that you must 
avoid making any sort of suggestions. All you need do 
is to appear completely ignorant, and even to make 
intentional mistakes so that the child may each time 
point out clearly what the rule is. Naturally, you must 
take the whole thing very seriously, all through the 
game. Then you ask who has won and why, and if 
everything is not quite clear, you begin a new set. 

It is of paramount importance during this first half 
of the interrogatory to play your part in a simple 
spirit and to let the child feel a certain superiority 
at the game (while not omitting to show by an occasional 
good shot that you are not a complete duffer) . In this way 
the child is put at his ease, and the information he gives as 
to how he plays is all the more conclusive. Many of our 
children became absorbed in the game to the extent of treat- 
ing me completely as one of them. " You are dished 1 " 
cries Ben (10 years) when my marble stops inside the square. 

In the case of the little ones, who find difficulty in 
formulating the rules which they observe in practice, 
the best way is to make them play in pairs. You begin 
by playing with one of them in the manner described 
above, and ask him to tell you all the rules he knows. 
Then you make the same request of the second boy 
(the first being no longer present), and finally you bring 
the two together and ask them to have a game. This 
control experiment is not needed for older children, except 
in doubtful cases. 

Then comes the second part of the interrogatory, that, 
namely, which bears upon the consciousness of rules. 
You begin by asking the child if he could invent a new 
rule. He generally does this easily enough, but it is 
advisable to make sure that it really is a new rule and 
not one of the many existing variants of which this 
particular child may already have knowledge. " I want 
a rule that is only by you, a rule that you've made up 


yourself and that no one else knows the rule of N 

(the child's name)/' Once the new rule has been formu- 
lated, you ask the child whether it could give rise to a 
new game : " Would it be all right to play like that 
with your pals ? Would they want to play that way ? 
etc." The child either agrees to the suggestion or disputes 
it. If he agrees, you immediately ask him whether the 
new rule is a " fair " rule, a " real " rule, one " like the 
others ", and try to get at the various motives that enter 
into the answers. If, on the other hand, the child dis- 
agrees with all this, you ask him whether the new rule, 
could not by being generalized become a real rule. " When 
you are a big boy, suppose you tell your new rule to a 
lot of children, then perhaps they'll all play that way 
and everyone will forget the old rules. Then which rule 
will be fairest yours that everyone knows, or the old 
one that everyone has forgotten ? " The formula can 
naturally be altered in accordance with the turn which 
the conversation is taking, but the main point is to find 
out whether one may legitimately alter rules and whether 
a rule is fair or just because it conforms to general usage 
(even newly introduced), or because it is endowed with 
an intrinsic and eternal value. 

Having cleared up this point it will be easy enough 
to ask the two following questions. (i) Have people 
always played as they do to-day : " Did your daddy 
play this way when he was little, and your grand-dad, 
and children in the time of William Tell, Noah, and Adam 
and Eve, etc., did they all play the way you showed me, 
or differently ? " (2) What is the origin of rules : Are 
they invented by children or laid down by parents and 
grown-ups in general ? 

Sometimes it is best to begin by these last two questions 
before asking whether rules can be changed ; this avoids 
perse ver at ion, or rather reverses its direction, and so 
facilitates the interpretation of the answers. All this 
part of the interrogatory, moreover, requires extremely 
delicate handling ; suggestion is always ready to occur, 


and the danger of romancing is ever present. But it 
goes without saying that the main thing is simply to 
grasp the child's mental orientation. Does he believe 
in the mystical virtue of rules or in their finality ? Does 
he subscribe to a heteronomy of divine law, or is he 
conscious of his own autonomy ? This is the only question 
that interests us. The child has naturally got no ready- 
made beliefs on the origin and endurance of the rules of 
his games ; the ideas which he invents then and there are 
only indices of his fundamental attitude, and this must 
be steadily borne in mind throughout the whole of the 

The results which we obtained from this double inter- 
rogatory and which we shall examine in greater detail 
later on, are roughly the following. 

From the point of view of the practice or application 
of rules four successive stages can be distinguished. 

A first stage of a purely motor and individual character, 
during which the child handles the marbles at the dictation 
of his desires and motor habits. This leads to the for- 
mation of more or less ritualized schemas, but since play 
is still purely individual, one can only talk of motor 
rules and not of truly collective rules. 

The second may be called egocentric for the following 
reasons. This stage begins at the moment when the 
child receives from outside the example of codified rules, 
that is to say, some time between the ages of two and 
five. But though the child imitates this example, he 
continues to play either by himself without bothering to 
find play-fellows, or with others, but without trying to 
win, and therefore without attempting to unify the 
different ways of playing. In other words, children of 
this stage, even when they are playing together, play 
each one " on his own " (everyone can win at once) 
and without regard for any codification of rules. This 
dual character, combining imitation of others with a 
purely individual use of the examples received, we have 
designated by the term Egocentrism, 


A third stage appears between 7 and 8, which we 
shall call the stage of incipient cooperation. Each player 
now tries to win, and all, therefore, begin to concern 
themselves with the question of mutual control and of 
unification of the rules. But while a certain agreement 
may be reached in the course of one game, ideas about 
the rules in general are still rather vague. In other 
words, children of 7-8, who belong to the same class 
at school and are therefore constantly playing with each 
other, give, when they are questioned separately, disparate 
and often entirely contradictory accounts of the rules 
observed in playing marbles. 

Finally, between the years of n and 12, appears a 
fourth stage, which is that of the codification of rules. 
Not only is every detail of procedure in the game fixed, 
but the actual code of rules to be observed is known to 
the whole society. There is remarkable concordance in 
the information given by children of 10-12 belonging to 
the same class at school, when they are questioned on the 
rules of the game and their possible variations. 

These stages must of course be taken only for what 
they are worth. It is convenient for the purposes of 
exposition to divide the children up in age-classes or 
stages, but the facts present themselves as a continuum 
which cannot be cut up into sections. This continuum, 
moreover, is not linear in character, and its general 
direction can only be observed by schematizing the 
material and ignoring the minor oscillations which render 
it infinitely complicated in detail. So that ten children 
chosen at random will perhaps not give the impression 1 
of a steady advance which gradually emerges from the 
interrogatory put to the hundred odd subjects examined 
by us at Geneva and NeucMtel. 

If, now, we turn to the consciousness of rules we shall 
find a progression that is even more elusive in detail, 
but no less clearly marked if taken on a big scale. We 
may express this by saying that the progression runs 
through three stages, of which the second begins during 


the egocentric stage and ends towards the middle of the 
stage of cooperation (9-10), and of which the third covers 
the remainder of this co-operating stage and the whole 
of the stage marked by the codification of rules. 

During the first stage rules are not yet coercive in 
character, either because they are purely motor, or else 
(at the beginning of the egocentric stage) because they 
are received, as it were, unconsciously, and as interesting 
examples rather than as obligatory realities. 

During the second stage (apogee of egocentric and 
first half of cooperating stage) rules are regarded as 
sacred and untouchable, emanating from adults and 
lasting forever. Every suggested alteration strikes the 
child as a transgression. 

Finally, during the third stage, a rule is looked upon 
as a law due to mutual consent, which you must respect 
if you want to be loyal but which it is permissible to 
alter on the condition of enlisting general opinion on 
your side. 

The correlation between the three stages in the develop- 
ment of the consciousness of rules and the four stages 
relating to their practical observance is of course only a 
statistical correlation and therefore very crude. But 
broadly speaking the relation seems to us indisputable. 
The collective rule is at first something external to the 
individual and consequently sacred to him ; then, as he 
gradually makes it his own, it comes to that extent to 
be felt as the free product of mutual agreement and an 
autonomous conscience. And with regard to practical 
use, it is only natural that a mystical respect for laws 
should be accompanied by a rudimentary knowledge and 
application of their contents, while a rational and well- 
founded respect is accompanied by an effective application 
of each rule in detail 

There would therefore seem to be two types of respect 
for rules corresponding to two types of social behaviour. 
This conclusion deserves to be closely examined, for if 
it holds good, it should be of the greatest value to the 


analysis of child morality. One can see at once all that 
it suggests in regard to the relation between child and 
adult. Take the insubordination of the child towards 
its parents and teachers, joined to its sincere respect for 
the commands it receives and its extraordinary mental 
docility. Could not this be due to that complex of 
attitudes which we can observe during the egocentric 
stage and which combines so paradoxically an unstable 
practice of the law with a mystical attitude towards it ? 
And will not cooperation between adult and child, in so 
far as it can be realized and in so far as it is facilitated 
by co-operation between children themselves, supply the 
key to the interiorization of commands and to the auto- 
nomy of, the moral consciousness ? Let us therefore not 
be afraid of devoting a certain amount of time to the 
patient analysis of the rules of a game, for we are here 
in possession of a method infinitely more supple, and 
consequently more sure, than that of merely questioning 
children about little stories, a method which we shall be 
obliged to adopt in the latter part of this book. 


STAGES. We need not dwell at any length upon the 
first stage, as it is not directly connected with our subject. 
At the same time, it is important that we should know 
whether the rules which come into being previous to any 
collaboration between children are of the same type as 
collective rules. 

Let us give a handful of ten marbles to a child of three 
years and four months and take note of its reactions : 

Jacqueline has the marbles in her hands and looks 
at them with curiosity (it is the first time she has seen 
any) ; then she lets them drop on to the carpet. After 
this she puts them in the hollow of an arm-chair. 
" Aren't they animals? Oh, no. Are they balk? Yes/' 
She puts them back on the carpet and lets them drop 
from a certain height. She sits on the carpet with her 
legs apart and throws the marbles a few inches in front 


of her. She then picks them up and puts them on the 
arm-chair and in the same hole as before. (The arm- 
chair is studded with buttons which create depressions 
in the material.) Then she collects the lot and lets them 
drop, first all together, then one by one. After this she 
replaces them in the aim-chair, first in the same place 
and then in the other holes. Then she piles them up 
in a pyramid : " What are marbles ? What do you 
think ?..." She puts them on the floor, then back 
on to the arm-chair, in the same holes. We both go 
out on to the balcony : she lets the marbles drop from a 
height to make them bounce, 

The following days, Jacqueline again places the marbles 
on the chairs and arm-chairs, or puts them into her 
little sauce-pan to cook dinner. Or else she simply repeats 
the behaviour described above. 

Three points should be noted with regard to facts 
such as these. In the first place, the lack of continuity 
and direction in the sequence of behaviour. The child 
is undoubtedly trying first and foremost to understand 
the nature of marbles and to adapt its motor schemas to 
this novel reality. This is why it tries one experiment 
after another : throwing them, heaping them into pyramids 
or nests, letting them drop, making them bounce, etc. 
But once it has got over the first moments of astonish- 
ment, the game still remains incoherent, or rather still 
subject to the whim of the moment. On days when the 
child plays at cooking dinner, the marbles serve as food 
to be stewed in a pot. On days when it is interested in 
classifying and arranging, the marbles are put in heaps 
in the holes of arm-chairs, and so on. In the general 
manner in which the game is carried on there are 
therefore no rules. 

The second thing to note is that there are certain 
regularities of detail, for it is remarkable how quickly 
certain particular acts in the child's behaviour become 
schematized and even ritualized. The act of collect- 
ing the marbles in the hollow of an arm-chair is at 
first simply an experiment, but it immediately becomes 
a motor schema bound up with the perception of the 


marbles. After a few days it is merely a rite, still per- 
formed with interest, but without any fresh effort of 

In the third place, it is important to note the symbolism 1 
that immediately becomes grafted upon the child's motor 
schemas. These symbols are undoubtedly enacted in play 
rather than thought out, but they imply a certain amount 
of imagination : the marbles are food to be cooked, eggs 
in a nest, etc. 

This being so, the rules of games might be thought to 
derive either from rites analogous to those we have just 
examined or from a symbolism that has become collective. 
Let us briefly examine the genesis and ultimate 'destiny 
of these modes of behaviour. 

Genetically speaking, the explanation both of rites and 
of symbols would seem to lie in the conditions of pre- 
verbal motor intelligence. When it is presented with 
any new thing, a baby of 5 to 8 months will respond with 
a dual reaction; it will accommodate itself to the new 
object and it will assimilate the object to earlier motor 
schemas. Give the baby a marble, and it will explore 
its surface and consistency, but will at the same time 
use it as something to grasp, to suck, to rub against the 
sides of its cradle, and so on. This assimilation of every 
fresh object to already existing motor schemas may be 
conceived of as the starting point of ritual acts and 
symbols, at any rate from the moment that assimilation 
becomes stronger than actual accommodation itself. With 
regard to ritual acts, indeed, one is struck by the fact 
that from the age of about 8 to 10 months all the child's 
motor schemas, apart from moments of adaptation in 
the real sense, give rise to a sort of functioning in the 
void, in which the child takes pleasure as in a game. 
Thus, after having contracted the habit of pressing her face 

1 We use the term " symbol " in the sense given to it in the linguistic 
school of Saussure, as the contrary of sign. A sign is arbitrary, a 
symbol is motivated. It is in this sense, too, that Freud speaks of 
symbolic thought. 


against her parents' cheeks, crumpling up her nose and 
breathing deeply the while, Jacqueline began to perform 
this rite as a joke, crumpling up her nose and breathing 
deeply in advance, merely suggesting contact with another 
person's face, but without, as before, expressing any 
particular affection by the act. Thus from being actual, 
and incorporated in an effective adaptation this schema 
has become ritualized and serves only as a game. 1 Or 
again, Jacqueline in her bath is engaged in rubbing her 
hair ; she lets go of it to splash the water. Immediately, 
she repeats the movement, touching her hair and the 
water alternately, and during the next few days the 
schema has become ritualized to such an extent that she 
cannot strike the surface of the water without first out- 
lining the movement of smoothing her hair. 2 In no way 
automatic, this rite is a game that amuses her by its 
very regularity. Anyone observing a baby of 10 to 
12 months will notice a number of these rites which 
undoubtedly anticipate the rules of future games. 

As for symbols, they appear towards the end of the 
first year and in consequence of the ritual acts. For the 
habit of repeating a given gesture ritually, gradually leads 
to the consciousness of " pretending ". The ritual of 
going to bed, for instance (laying down one's head and 
arranging the corner of the pillow with the hundred and 
one complications which every baby invents), is sooner 
or later utilized " in the void ", and the smile of the 
child as it shuts its eyes in carrying out this rite is enough 
to show that it is perfectly conscious of " pretending " 
to go to sleep. Here already we have a symbol, but a 
" played " symbol. Finally, when language and imagery 
come to be added to motor intelligence, the symbol 
becomes an object of thought. The child who pushes a 
box along saying " tuff-tuff " is assimilating in imagination 
the box's movement to that of a motor-car : the play 
symbol has definitely come into being. 

This being so, can one seek among rites and symbols 

1 Age : 10 months. a Age : 12 months. 


for the origin of the actual rules of games ? Can the 
game of marbles, with its infinite complexity both with 
regard to the actual rules and to all that relates to the 
verbo-motor system of signs in use can the game of 
marbles, then, be conceived simply as the result of an 
accumulation of individual rites and symbols ? We do 
not think that it can. We believe that the individual 
rite and the individual symbol constitute the substructure 
for the development of rules and collective signs, its 
necessary, but not its sufficient condition. There is 
something more in the collective rule than in the motor 
rule or the individual ritual, just as there is something 
more in the sign than in the symbol. 

With regard to motor or ritualistic rules, there can be 
no doubt that they have something in common with 
rules in the ordinary sense, namely the consciousness of 
regularity. When we see the delight taken by a baby 
of 10 to 12 months or a child of 2-3 in reproducing a 
given behaviour in all its details, and the scrupulous 
attention with which it observes the right order in these 
operations, we cannot help recognizing the Regelbewusstsein 
of which Biihler speaks. But we must distinguish care- 
fully between the behaviour into which there enters only 
the pleasure of regularity, and that into which there 
enters an element of obligation. It is this consciousness 
of obligation which seems to us, as to Durkheim 1 and 
Bovet, 2 to distinguish a rule in the true sense from mere 

Now this element of obligation, or, to confine ourselves 
to the question of the practice of rules, this element of 
obedience intervenes as soon as there is a society, i.e. a 
relation between at least two individuals. As soon as a 
ritual is imposed upon a child by adults or seniors for whom 
he has respect (Bovet), or as soon, we would add, as a 
ritual comes into being as the result of the collaboration 

1 L' Education Morale. 

2 " Les Conditions de 1' Obligation de la Conscience *', Annie PsychoL 


of two children, it acquires in the subject's mind a 
new character which is precisely that of a rule. This 
character may vary according to the type of respect 
which predominates (respect for the senior or mutual 
respect) but in all cases there enters an element of sub- 
mission which was not contained in the rite pure and 

In actual fact, of course, there is every degree of variety 
between the simple regularity discovered by the individual 
and the "rule to which a whole social group submits itself. 
Thus during the egocentric stage we can observe a whole 
series of cases in which the child will use a rule as a mere 
rite, to be bent and modified at will, while at the same 
time he already tries to submit to the common laws. 
Just as the child very soon acquires the use of language 
and of the abstract and general concepts while retaining 
in his attitude to these much that still belongs to ego- 
centric modes of thought and even to the methods 
peculiar to symbolic and play thought, so, under the rules 
that are imposed upon him, he will for a long time contrive 
(in all good faith, needless to say) to maintain his own 
phantasy in the matter of personal decisions. But this 
factual continuity between ritual and rule does not exclude 
a qualitative difference between the two types of behaviour. 

Let us not, however, anticipate what will be said in 
our analysis of the consciousness of rules, but return to 
the matter of ritual. The individual rite develops quite 
naturally, as we have just shown, into a more or less 
complex symbolism. Can this symbolism be regarded as 
the starting point of that system of obligatory verbo- 
motor signs which are connected with the rules of every 
collective game ? As with the previous problem, we 
believe that the symbol is a necessary, but not a sufficient 
condition of the appearance of signs. The sign is general 
and abstract (arbitrary), the symbol is individual and 
motivated. If the sign is to follow upon the symbol, a 
group must therefore strip the individual's imagination 
of all its personal fantasy and then elaborate a common 


and obligatory imagery which will go hand in hand with 
the code of rules itself. 

Here is an observation showing how far removed are 
individual rites and symbols from rules and signs, though 
moving towards these realities in so far as collaboration 
between children becomes established. 

Jacqueline (after the observations given above) is 
playing with Jacques (2 years, n months and 15 days), 
who sees marbles for the first time. I. Jacques takes the 
marbles and lets them drop from a height one after 
another. After which he picks them up and goes away, 
II. Jacques arranges them on the ground, in a hollow 
and says, " Tm making a little nest ". Jacqueline takes 
one and sticks it in the ground in imitation. III. Jacques 
also takes one, buries it and makes a mud-pie above it. 
He digs it up and begins over again. Then he takes 2 at 
a time which he buries. Then 3, 4, 5 and up to 6 at a 
time, increasing the number of marbles systematically 
each time by one. Jacqueline imitates him : she first 
puts one marble down and makes a mud-pie over it, 
then two or three at random and without adopting a 
fixed system of progression. IV. Jacques puts all the 
marbles on a pile, then he places an india-rubber ball 
beside them and says : '* That's the Mummy ball and the 
baby balls." V. He piles them together again and covers 
them up with earth which he levels down. Jacqueline 
imitates him but with only one marble which she covers 
up without levelling the earth. She adds : " It's lost ", 
then digs it up and begins over again. 

This example shows very clearly how all the elements 
of individual fantasy or symbolism remain uncom- 
municated ; as soon as the game takes on an imaginative 
turn each child evokes its favourite images without 
paying any attention to anyone else's. It will also be 
observed how totally devoid of any general direction are 
the ritualized schemas successively tried. But as soon 
as there is reciprocal imitation (end of II and whole 
of III) we have the beginnings of a rule : each child 
tries to bury the marbles in the same way as the other, 
in a common order only more or less successfully adhered 
to. In bringing out this aspect, the observation leads us 


to the stage of egocentrism during which the child learns 
other peoples' rules but practises them in accordance 
with his own fantasy. 

We shall conclude this analysis of the first stage by 
repeating that before games are played in common, no 
rules in the proper sense can come into existence. 
Regularities and ritualized schemas are already there, 
but these rites, being the work of the individual, cannot 
call forth that submission to something superior to the 
self which characterizes the appearance of any rule. 

The second stage is the stage of egocentrism. In 
studying the practice of rules we shall make use of a 
notion which has served on earlier occasions in the 
descriptions we have given of the child's intellectual 
behaviour ; and, in both cases, indeed, the phenomenon is 
of exactly the same order. Egocentrism appears to us as a 
form of behaviour intermediate between purely individual 
and socialized behaviour. Through imitation and language, 
as also through the whole content of adult thought which 
exercises pressure on the child's mind as soon as verbal 
intercourse has become possible, the child begins, in a 
sense, to be socialized from the end of its first year. But 
the very nature of the relations which the child sustains 
with the adults around him prevents this socialization 
for the moment from reaching that state of equilibrium 
which is propitious to the development of reason. We 
mean, of course, the state of cooperation, in which the 
individuals, regarding each other as equals, can exercise 
a mutual control and thus attain to objectivity. In 
other words, the very nature of the relation between 
child and adult places the child apart, so that his thought 
is isolated, and while he believes himself to be sharing 
the point of view of the world at large he is really still 
shut up in his own point of view. The social bond itself, 
by which the child is held, close as it may seem when 
viewed from outside, thus implies an unconscious in- 
tellectual egocentrism which is further promoted by the 
spontaneous egocentrism peculiar to all primitive mentality. 


Similarly, with regard to the rules of games, it is easy 
to see, and greater authorities than ourselves 1 have 
already pointed out that the beginnings of children's 
games are characterized by long periods of egocentrisra. 
The child is dominated on the one hand by a whole set 
of rules and examples that are imposed upon him from 
outside. But unable as he is, on the other hand, to place 
himself on a level of equality with regard to his seniors, 
he utilizes for his own ends, unaware even of his own 
isolation, all that he has succeeded in grasping of the 
social realities that surround him. 

To confine ourselves to the game of marbles, the child 
of 3 to 5 years old will discover, according to what other 
children he may happen to come across, that in order to 
play this game one must trace a square, put the marbles 
inside it, try to expel the marbles from the square by 
hitting them with another marble, start from a line 
Jhat has been drawn beforehand, and so on. But though 
he imitates what he observes, and believes in perfect 
good faith that he is playing like the others, the child 
thinks of nothing at first but of utilizing these new 
acquisitions for himself. He plays in an individualistic 
manner with material that is social. Such is egocentrism. 

Let us analyse the facts of the case. 

MAR (6) 2 seizes hold of the marbles we offer him, and 
without bothering to make a square he heaps them up 
together and begins to hit the pile. He removes the 
marbles he has displaced and puts them aside or re- 
places them immediately without any method. " Do you 
always play like that ? In the street you make a square. 
WeU, you do the same as they do in the street. I'm 
waking a square, I am." (He ~draws the square, places 
the marbles inside it and begins to play again.) I play 

1 Stern in his Psychology of Early Childhood notes the identity of 
the stages we have established in children's conversations with those 
he has himself established with regard to play, pp. 177 and 332. 

2 The numbers in brackets give the child's age. The words of the 
child are in italics, those of the examiner in Roman lettering. Inverted 
commas mark the beginning and end of a conversation reported 
verbatim. All the subjects are boys unless the letter G. is added, 
indicating that the subject is a girl. 


with him, imitating each of his movements. " Who has 
won? We've both won. But who has won most?. . . " 
(Mar does not understand.) 

BAUM (61) begins by making a square and puts down 
three marbles, adding : " Sometimes you put 4, or 3, or 2. 

Or 5 ? No, not 5, but sometimes 6 or 8. Who begins 

when you play with the boys ISometimes me, sometimes 
the other one. Isn't there a dodge for knowing who is 
to begin? No. Do you know what a coche is? 
Rather ! " But the sequel shows that he knows nothing 
about the coche and thinks of this word as designating 
another game. " And which of us will begin ? You. 
Why ? / want to see how you do it." We play for a 
while and I ask who has won : " The one who has hit a 
mib? well, he has won.Wett ! who has won iI have, 
and then you' 1 I then arrange things so as to take 4 
while he takes 2 : " Who has won ? I have, and then 
you." We begin again. He takes two, I none. " Who 
has won ? I have. And I ? You've lost: 1 

LOEFF (6) often pretends to be playing with Mae, of 
whom we shall speak later. He knows neither how to 
make a square nor to draw a coche. He immediately 
begins to "fire" at the marbles assembled in a heap 
and plays without either stopping or paying any attention 
to us. "Have you won? I don't know. / think I 
h avet wiry ? Yes t because I threw the mils. 1 And I ? 
Yes, because you threw the mibs." 

DESARZ (6) : " Do you play often ? Yes, rather ! 
With whom IAll by myself. Do you like playing alone 
best ? You don't need two. You can play only one." 
He gathers the marbles together without a square and 
" fires " into the heap. 

Let us now see how two children, who have grown 
accustomed to playing together, set about it when they 
are left alone. They are two boys of whom one (Mae) 
is a very representative example of the present stage, 
while the other (Wid) stands at the border line between 
the present stage and the next. The analyses of these 
cases will be all the more conclusive as the children in 
question are no mere beginners at the game. 

MAE (6) and WID (7) declare that they are always 
playing together. Mae tells us that they both "played 

1 English equivalent for " marbre ". [Trans.] 


again, yesterday ". I first examine Mae by himself. 
He piles Ms marbles in a corner without counting them 
and throws his shooter into the pile. He then places 
4 marbles close together and puts a fifth on top (in a 
pyramid). Mae denies that a square is ever drawn. 
Then he corrects himself and affirms that he always does 
so : " How do you and Wid know which is to begin ? 
One of the two throws his shooter and the other tries to hit 
it. If he hits it, he begins." Mae then shows us what 
the game consists in : he throws his shooter without 
taking into account the distances or the manner of playing 
("piquette"), and when he succeeds in driving a marble 
out of the square he immediately puts it back. Thus the 
game has no end. " Does it go on Hke that all the time ? 
You take one away to make a change (he takes a marble out 
of the square, but not the one that he has touched). It'll 
only "be finished when there's only one left (he ' fires * again 
twice). One more shot, and then you take one away." 
Then he affirms : " every third shot you take one away" 
He does so. Mae removes a marble every third shot 
independently of whether he has hit or missed, which is 
completely irregular and corresponds to nothing in the 
game as habitually played, or as we have seen it played 
in Neuchatel or Geneva. It is therefore a rule which 
he has invented then and there but which he has the 
impression of remembering because it presents a vague 
analogy with what really happens when the player removes 
the marble he has just " hit " (touched) . This game of Mae's 
is therefore a characteristic game of the second stage, an 
egocentric game in which " to win " does not mean getting 
the better of the others, but simply playing on one's own. 
Wid, whom I now prepare to question and who 
has not assisted at Mae's interrogatory, begins by making 
a square. He places 4 marbles at the 4 corners and one 
in the middle (the same disposition as Mae's, which was 
probably a deformation of it). Wid does not know what 
to do to decide which is to begin, and declares that he 
understands nothing of the method which Mae had 
shewn me as being familiar to both of them (trying to hit 
one's partner's shooter). Wid then throws his shooter in 
the direction of the square, knocking out one marble 
which he puts in his pocket. Then I take my turn, 
but fail to touch anything. He plays again and wins 
all the marbles, one after the other, keeping them each 
time. He also declares that when you have knocked a 


marble out, you have the right to play another shot 
straight away. After having taken everything he says : 
" I've won. 39 Wid therefore belongs to the third stage 
if this explanation is taken as a whole, but the sequel 
will show that he takes no notice of Mae's doings when 
they are playing together. Wid stands therefore at the 
boundary line which separates the stage of egocentrism 
from the stage of cooperation. 

I then tell Mae to come into the room and the two 
children begin to play with each other. Mae draws a 
square and Wid disposes the marbles in accordance with 
Ms habitual schema, Mae begins (he plays " Roulette " 
whereas Wid most of the time plays " Piquette ") and 
dislodges four marbles. " I can play four times, now ", 
adds Mae. This is contrary to all the rules, but Wid 
finds the statement quite natural. So one game succeeds 
another. But the marbles are placed in the square by 
one child or the other as the spirit moves them (according 
to the rules each must put his " pose ") and the dis- 
lodged marbles are sometimes put straight back into the 
square, sometimes retained by the boy who has won 
them. Each plays from whatever place he chooses, 
unchecked by his partner, and each " fires " as many 
times as he likes (it thus often happens that Mae and 
Wid are playing at the same time). 

I now send Wid out of the room and ask Mae to 
explain the game to us for a last time. Mae places 16 
marbles in the middle of the square. " Why so many 
as that ? So as to win. How many do you put down 
at home with Wid ? / put five, but when I'm alone, I 
put lots." Mae then begins to play and dislodges a marble 
which he puts on one side. I do the same. The game 
continues in this way, each playing one shot at a time 
without taking the dislodged marbles into account (which 
is contrary to what Mae was doing a moment ago). Mae 
then places five marbles in the square, like Wid. This 
time I arrange the five marbles as Mae himself had 
done at the beginning of the interrogatory (four close 
together and one on top) but Mae seems to have forgotten 
this way of doing things. In the end Mae plays by taking 
away a marble every three shots, as before, and says to 
us : " It's so that it should stop" 

We have quoted the whole of this example in order 
to show how little two children from the same class at 


school, living in the same house, and accustomed to 
playing with each other, are able to understand each 
other at this age. Not only do they tell us of totally 
different rules (this still occurs throughout the third 
stage), but when they play together they do not watch 
each other and do not unify their respective rules even 
for the duration of one game. The fact of the matter 
is that neither is trying to get the better of the other : 
each is merely having a game on Ms own, trying to hit 
the marbles in the square, i.e. trying to " win " from his 
own point of view. 

This shows the characteristics of the stage. The child 
plays for himself. His interest does not in any way 
consist in competing with his companions and in binding 
himself by common rules so as to see who will get the 
better of the others. His aims are different. They are 
indeed dual, and it is this mixed behaviour that really 
defines egocentrism. On the one hand, the child feels 
very strongly the desire to play like the other boys, 
and especially like those older than himself; he longs, 
that is to say, to feel himself a member of the very 
honourable fraternity of those who know how to play 
marbles correctly. But quickly persuading himself, on 
the other hand, that his playing is " right " (he can 
convince himself as easily on this point as in all his 
attempts to imitate adult behaviour) the child thinks 
only of utilizing these acquisitions for himself : Ms 
pleasure still consists in the mere development of skill, 
in carrying out the strokes he sets himself to play. It 
is, as in the previous stage, essentially a motor pleasure, 
not a social one. The true " socius " of the player who 
has reached this stage is not the flesh and blood partner 
but the ideal and abstract elder whom one inwardly 
strives to imitate and who sums up all the examples one 
has ever received. 

It little matters, therefore, what one's companion is 
doing, since one is not trying to contend against him. 
It little matters what the details of the rules may be, 


since there is no real contact between the players. This 
is why the child, as soon as he can schematically copy 
the big boys* game, believes himself to be in possession 
of the whole truth. Each for himself, and all in com- 
munion with the " Elder " : such might be the formula 
of egocentric play. 

It is striking to note the affinity between this attitude 
of children of 4 to 6 in the game of marbles and the 
attitude of those same children in their conversations 
with each other. For alongside of the rare cases of true 
conversation where there is a genuine interchange of 
opinions or commands, one can observe in children 
between 2 and 6 a characteristic type of pseudo-conver- 
sation or " collective monologue ", during which the 
children speak only for themselves, although they wish 
to be in the presence of interlocutors who will serve as 
a stimulus. Now here again, each feels himself to be in 
communion with the group because he is inwardly address- 
ing the Adult who knows and understands everything, 
but here again, each is only concerned with himself, for 
lack of having dissociated the "ego " from the " socius ". 

These features of the egocentric stage will not, how- 
ever, appear in their full light until we come to analyse 
the consciousness of rules which accompanies this type 
of conduct. 


STAGES. Towards the age of 7-8 appears the desire for 
mutual understanding in the sphere of play (as also, 
indeed, in the conversations between children). This 
felt need for understanding is what defines the third 
stage. As a criterion of the appearance of this stage we 
shall take the moment when by " winning " the child 
refers to the fact of getting the better of the others, 
therefore of gaining more marbles than the others, and 
when he no longer says he has won when he has done no 
more than to knock a marble out of the square, regardless 
of what his partners have done. As a matter of fact, 


no child, even from among the older ones, ever attributes 
very great importance to the fact of knocking out a few 
more marbles than Ms opponents. Mere competition Is 
therefore not what constitutes the affective motive-power 
of the game. In seeking to win the child is trying 
above all to contend with his partners while observing 
common rules. The specific pleasure of the game thus 
ceases to be muscular and egocentric, and becomes social. 
Henceforth, a game of marbles constitutes the equivalent 
in action of what takes place in discussion in words : 
a mutual evaluation of the competing powers which 
leads, thanks to the observation of common rules, to a 
conclusion that is accepted by all. 

As to the difference between the third and fourth 
stages, it is only one of degree. The children of about 
7* to 10 (third stage) do not yet know the rules in detail. 
They try to learn them owing to their increasing interest 
in the game played in common, but when different 
children of the same class at school are questioned on the 
subject the discrepancies are still considerable in the 
information obtained. It is only when they are at play 
that these same children succeed in understanding each 
other, either by copying the boy who seems to know 
most about it, or, more frequently, by "omitting any 
usage that might be disputed. In this way they play 
a sort of simplified game. Children of the fourth stage, 
on the contrary, have thoroughly mastered their code 
and even take pleasure in juridical discussions, whether 
of principle or merely of procedure, which may at times 
arise out of the points in dispute. 

Let us examine some examples of the third stage, and, 
in order to point more clearly to the differentiating 
characters of this stage, let us begin by setting side by 
side the answers of two little boys attending the same 
class at school and accustomed to playing together. 
(The children were naturally questioned separately in 
order to avoid any suggestion between them, but we 
afterwards compared their answers with one another.) 


BEN (10) and Nus (n, backward, one year below the 
school standard) are both in the fourth year of the lower 
school and both play marbles a great deal. They agree 
in regarding the square as necessary. Nus declares that 
you always place 4 marbles in the "square, either at the 
corners or else 3 in the centre with one on top (in a 
pyramid). Ben, however, tells us that you place 2 to 10 
marbles in the enclosure (not less than 2, not more than 10). 

To know who is to begin you draw, according to Nus, 
a line called the " coche " and everyone tries to get near 
it : whoever gets nearest plays first, and whoever goes 
beyond it plays last. Ben, however, knows nothing 
about the coche : you begin " as you like. Isn't there a 
dodge for knowing who is to play first ? No. Don't 
you try with the coche ? Yes, sometimes. What is the 
coche ? . . . (he cannot explain)/' On the other hand, 
Ben affirms that you " fire " the first shot at a distance 
of 2 to 3 steps from the square. A single step is not 
enough, and "four isn't any good either ". Nus is ignorant 
of this law and considers the distance to be a matter of 

With regard to the manner of " firing ", Nus is equally 
tolerant. According to him you can play " piquette " 
or " roulette ", but " when you play piquette everyone 
must play the same. When one boy says that you must 
play roulette, everyone plays that way ". Nus prefers 
roulette because " that is the best way 31 : piquette is 
more difficult. Ben, however, regards piquette as 
obligatory in all cases. He is ignorant, moreover, of the 
term roulette and when we show him what it is he says : 
" That is bowled piquette/ [Fr., Piquette roulee] That's 
cheating I " 

According to Nus everyone must play from the coche, 
and aU through the game. When, after having shot at 
the square you land anywhere, you must therefore come 
back to the coche to " fire " the next shot. Ben, on 
the contrary, who on this point represents the more 
genera] usage, is of opinion that only the first shot should 
be fired from the coche : after that " you must play from 
where you are." 

Nus and Ben thus agree in stating that the marbles 
that have gone out of the square remain in the possession 
of ^ the boy who dislodged them. This is the only point, 
this and the actual drawing of the square, on which the 
children give us results that are in agreement. 


When we begin to play, I arrange to stay in the square 
(to leave my shooter inside the enclosure). " You are 
dished (Fr. cuit), cries Ben,, delighted, you can't play 
again until I get you out ! " Nus knows nothing of this 
rule. Again, when I play carelessly and let the shooter 
drop out of my hand, Ben exclaims " Fan-coup " to 
prevent me from saying " coup-passe " and having another 
shot. Nus is ignorant of this rule. 

At one point Ben succeeds in hitting my shooter. 
He concludes from this that he can have another shot, 
just as though he had hit one of the marbles placed in 
the square. Nus, in the same circumstances does not 
draw the same conclusions (each must play in turn 
according to him) but deduces that he wUl be able to 
play the first shot in the next game. 

In the same way, Ben thinks that everyone plays from 
the place the last shot has led him to and knows the rule 
that authorizes the player to change places, saying " du 
mien " or " un empan ", whereas Nus, who has certainly 
heard those words, does not know what they mean. 

These two cases, chosen at random out of a class of 
lo-year-old pupils, show straight away what are the two 
differential features of the second stage. i There is 
a general will to discover the rules that are fixed and 
common to all players (cf. the way Nus explains to us 
that if one of the partners plays piquette " everyone 
must play the same "). 2 In spite of this there is con- 
siderable discrepancy in the children's information, Lest 
the reader should think the above examples exceptional 
here are, on the same point, the answers of another child 
from the same class : 

Ross (n ; i) : " First, every one puts two marbles on 
the square. You can make the square bigger when there 
are more playing." Ross knows the method of the coche 
for knowing who is to begin. Like Nus, he allows both 
roulette and piquette. He also allows what is not only con- 
trary to all established usages but also to the sense of the 
words, a way of playing which he calls " femme-poussette " 
which consists in carrying one's hand along with the 
marble as one throws it (push stroke in billiards). Now 
this is always forbidden, and the very word that Ross 


has deformed says so " fan-poussette ". According to 
Ross, you play from the place you have reached with the 
last shot, and when you have won a marble you have the 
right to another shot straight away. To change your place 
you must say " du mien ". " // a stone gets in our way, you 
say ' coup-pass i ' and have another shot. If it slips [if the 
marble slips out of your hand] you say ' laehi* (Engl. 
4 gone '). // you don't say that, you can't have another 
turn. It's the rules ! " Ross here stands mid-way between 
Nus and Ben. Finally, Ross knows of a rather ^ peculiar 
custom which is unknown to Nus and Ben. " If you 
stay in the square you can be hit and then he picks up the 
marbles [ = If your shooter stays inside the square and 
is touched by your opponent's shooter, he is entitled to 
all the marbles in the square]. He (the opponent) can 
have two shots [to try and hit the shooter in question] 
and if he misses tfa first he can take [at the second shot] 
the shooter from anywhere [though of course only from 
the outside of the square] and make the marbles go out 
[ = take them] ". This rule has generally only been des- 
cribed to us by children of the fourth stage, but the rest 
of Ross' interrogatory is typically third stage. 

Such then is the third stage. The child's chief interest 
is no longer psycho-motor, it is social. In other words, 
to dislodge a marble from a square by manual dexterity 
is no longer an aim in itself. The thing now is not only 
to fight the other boys but also and primarily to regulate 
the game with a whole set of systematic rules which will 
ensure the most complete reciprocity in the methods used. 
The game has therefore become social. We say " become " 
because it is only after this stage that any real cooperation 
exists between the players. Before this, each played for 
himself. Each sought, it is true, to imitate the play of 
older boys and of the initiated, but more for the satis- 
faction, still purely personal, of feeling himself to be a 
member of a mystical community whose sacred institutions 
are handed down by the elders out of the remote past, 
than from any real desire to cooperate with his playmates 
or with anyone else. If cooperation be regarded as more 
social than this mixture of egocentrism and respect for 
one's seniors which characterizes the beginnings of 


collective life among children, then we may say that it 
is from the third stage onwards that the game of marbles 
begins to be a truly social game. 

As yet, however, tMs cooperation exists to a great 
extent only in intention. Being an honest man is not 
enough to make one to know the law. It is not even 
enough to enable us to solve all the problems that may 
arise in our concrete "moral experience ". The child fares 
in the same way during the present stage, and succeeds, 
at best, in creating for himself a " provisional morality ", 
putting off till a later date the task of setting up a code of 
laws and a system of jurisprudence. Nor do boys of 7 to 
10 ever succeed in agreeing amongst themselves for longer 
than the duration of one and the same game ; they are 
still incapable of legislating on all possible cases that 
may arise, for each still has a purely personal opinion 
about the rules of the game. 

To use an apter comparison, we may say that the child 
of 7 to 10 plays as he reasons. We have already L tried 
to establish the fact that about the age of 7 or 8, precisely, 
that is to say, at the moment when our third stage appears, 
in the very poor districts where we conducted our work, 2 
discussion and reflection gain an increasing ascendency 
over unproved affirmation and intellectual egocentrism. 
Now, these new habits of thought lead to genuine 
deductions (as opposed to primitive " transductions ") 
and to deductions in which the child grapples with a 
given fact of experience, either present or past. But 
something is still lacking if deduction is to be generalized 
and made completely rational : the child must be able 
to reason formally, i.e. he must have a conscious realization 
of the rules of reasoning which will enable him to apply 
them to any case whatsoever, including purely hypo- 

* /.#., chap. IV. 

* We take this opportunity of reminding the reader of what has not 
been sufficiently emphasized in our earlier books, viz. that most of 
our research has been carried out on children from the poorer parts 
of Geneva. In different surroundings the age averages would certainly 
have been different. 


thetical cases (mere assumptions). In the same way, a 
child who, with regard to the rules of games, has reached 
the third stage, will achieve momentary coordinations 
of a coEective order (a well ordered game may be com- 
pared on this point to a good discussion), but feels no 
interest as yet in the actual legislation of the game, in 
the discussions of principle which alone will give him 
complete mastery of the game in all its strictness. (From 
this point of view the juridico-moral discussions of the 
fourth stage may be compared to formal reasoning in 

It is, on an average, towards the age of u or 12 that 
these interests develop. In order to understand what is 
the practice of rules among children of this fourth stage 
let us question separately several children from the same 
class at school, and we shall see how subtle are their 
answers, and how well they agree with one another. 

RIT (12), GROS (13) and VUA (13) often play marbles. 
We questioned them each separately and took steps to 
prevent them from communicating to each other during 
our absence the contents of our interrogatory. 

With regard to the square, the " pose ", the manner 
of throwing, and generally speaking all the rules we 
have already examined, these three children are naturally 
in full agreement with each other. To know who is to 
play first, Rit, who has lived in two neighbouring villages 
before coming to town, tells us that various customs are 
in usage. You draw a line, the coche, and whoever gets 
nearest to it plays first. If you go beyond the line, 
either, according to some, it does not matter, or else 
" there is another game : when you go beyond the line, you 
play last 3 '. Gros and Vua know only of this custom, 
the only one that is really put into practice by the boys 
of the neighbourhood. 

But there are complications about which the younger 
boys left us in the dark. " Whoever, according to Gros, 
says ' queue ' plays second. It's easier because he doesn't 
get^ ' hit ' [ = if a player's shooter lands near the square, 
it is exposed to hits from the other players]." In the same 
way, Vua tells us that " whoever says ' queue de deux ' 
plays last ' ' . And he adds the following rule, also recognized 


by Gros : * ' When you are all at the same distance from the 
cache whoever cries ' egaux-queue ' plays second " (the 
problem is therefore to play sufficiently soon still to 
find marbles in the square, but not first, for fear of 
being hit). 

On the other hand, Gros tells us : " Whoever takes out 
two [two of the marbles placed inside the square, i.e. the 
equivalent of the player's ' pose '] can say ' queue-de- 
pose '. In that way he can play second from the coche in 
the next game." And Vua : " When there are two outside 
[when two marbles have been knocked out of the square] 
you can dare to say ' queue-de-pose ', and you can play 
second from the coche again in the second game. 13 Rit 
gives us the same information. 

This is not all. According to Rit, " if you say f deux- 
coups-de-coche ' you can have two shots from the line. If 
you say ' deux-coups-d* empan ' you play the second shot 
from where you are. You can only say that when the 
other [ = the opponent] has made up his pose [ = has won 
back as many marbles as he had originally deposited in 
the square] ". This rule is observed in the same way by 
the other two children. 

In addition, there is a whole set of rules, unknown 
to the younger boys, which bear upon the position of the 
marbles in the square. According to Gros " the first boy 
who says ' place-pour-moi * [Eng., place-for-me] does not 
have to place himself at one of the corners of the square ", 
and "the one who has said ' places-des-marbres* (Engl., 
place for the marbles) can put them down as he likes, in a 
' troy at ' (all in a heap) or at the four corners ". Vua is 
of the same opinion and adds : " If you say ' place-pour- 
toi-pour-tout-le-jeu (Engl., your-place-for-the-whole-game) 
the other chap [ = the opponent] must stay at the same 
place" Rit, who knows both these rules, adds the further 
detail that "you can't say 'place-pour-moi' if you have 
already said ' place-pour-toi V This gives some idea of 
the complications of procedure ! 

Our three legal experts also point the measures of 
clemency in use for the protection of the weak. According 
to Vua " if you knock out three at one shot and there s 
only one left [one marble in the square] the other chap 
[the opponent] has the right to play from half-way [half- 
way between the coche and the square] because the first 
boy has made more than his ' pose V Also : " the boy 
who has been beaten is allowed to begin. 1 ' According to 


Gros, " if there is one marble left at the end, the boy who 
has won, instead of taking it, can give to the other chap'' 
And again, ** When there's one boy who has won too much, 
the others say ' coufac ', and he is bound to play another 

The number of shots at the disposal of each player also 
gives rise to a whole series of regulations on which the 
three boys lay stress, as before, in full agreement with 
each other. For the sake of brevity we refer the reader 
on this point to the general rules outlined in Section I. 

There is only one point on which we saw our subjects 
differ. Rit who, it will be remembered, has known the 
game in three different districts, tells us that the boy 
whose shooter stays inside the square may generally 
come out of it. He added, it is true, that in some games 
the player in such a plight is " dished " (Fr., bride), but 
this rule does not seem to him obligatory. Vua and 
Gros, on the contrary, are of opinion that in all cases 
" when you stay inside the square you are dished ". We 
think we may confuse Vua by saying : " Rit didn't say 
that ! The fact is, answers Vua, that sometimes people 
play differently. Then you ask each other what you want 
to do. And 2 you can't agree ? We scrap for a bit and 
then we fix things up." 

These answers show what the fourth stage is. Interest 
seems to have shifted its ground since the last stage. 
Not only do these children seek to cooperate, to " fix 
things up/' as Vua puts it, rather than to play for them- 
selves alone, but also and this undoubtedly is something 
new they seem to take a peculiar pleasure in anticipating 
all possible cases and in codifying them. Considering 
that the square game is only one of the five or ten 
varieties of the game of marbles, it is almost alarming 
in face of the complexity of rules and procedure in the 
square game, to think of what a child of twelve has to 
store away in his memory. These rules, with their over- 
lapping and their exceptions, are at least as complex as 
the current rules of spelling. It is somewhat humiliating, 
in this connection, to see how heavily traditional education 
sets about the task of making spelling enter into brains 
that assimilate with such ease the mnemonic contents of 


the game of marbles. But then, memory is dependent 
upon activity, and a real activity presupposes interest. 

Throughout this fourth stage, then, the dominating 
interest seems to be interest in the rules themselves. For 
mere cooperation would not require such subtleties as 
those attending the disposition of the marbles in the 
square (" place-pour-moi ", " place-des-marbres ", " place- 
pour-toi-pour-tout-le-jeu ", etc.). The fact that the child 
enjoys complicating things at will proves that what he 
is after is rules for their own sake. We have described 
elsewhere l the extraordinary behaviour of eight boys of 
10 to ii who, in order to throw snow-balls at each other, 
began by wasting a good quarter-of-an-hour in electing 
a president, fixing the rules of voting, then in dividing 
themselves into two camps, in deciding upon the distances 
of the shots/ and finally in foreseeing what would be the 
sanctions to be applied in cases of infringement of these 
laws. Many other facts analogous to this could be culled 
from studies that have been made on children's societies. 

In conclusion, the acquisition and practice of the rules 
of a game follow very simple and very natural laws, the 
stages of which may be defined as follows : i Simple 
individual regularity. 2 Imitation of seniors with ego- 
centrism. 3 Cooperation. 4 Interest in rules for their 
own sake. Let us now see whether the consciousness of 
rules describes in its evolution an equally uncomplicated 

STAGES. As all our results have shown, consciousness of 
rules cannot be isolated from the moral life of the child 
as a whole. We might, at the most, study the practical 
applications of rules without bothering about obedience 
in general, i.e. about the child's whole social and moral 
behaviour. But as soon as we try, as in the present 
case, to analyse a child's feelings and thoughts about 
rules, we shall find that he assimilates them unconsciously 

i /.*., p. 96. 


along with the commands to which he is subjected taken 
as a whole. This comes out particularly clearly in the 
case of the little ones, for whom the constraint exercised 
by older children evokes adult authority itself in an 
attenuated form. 

Thus the great difficulty here, even more than with 
the practice of rules, is to establish the exact significance 
of the primitive facts. Do the simple individual regu- 
larities that precede the rules imposed by a group of 
players give rise to the consciousness of rules, or do they 
not ? And if they do, is this consciousness directly 
influenced by the commands of adults ? This very 
delicate point must be settled before we can embark 
upon the analysis of the more transparent data furnished 
by the interrogatory of older children. With regard to 
consciousness of rules, we shall designate as the first 
stage that which corresponds to the purely individualistic 
stage studied above. During this stage the child, as we 
noted, plays at marbles in its own way, seeking merely 
to satisfy its motor interests or its symbolic fantasy. 
Only, it very soon contracts habits which constitute 
individual rules of a sort.. This phenomenon, far from 
being unique, is the counterpart of that sort of ritualization 
of behaviour which can be observed in any baby before 
it can speak or have experienced any specifically moral 
adult pressure. Not only does every act of adaptation 
extend beyond its content of intellectual effort into a 
ritual kept up for its own sake, but the baby will often 
invent such rituals for its own pleasure ; hence the 
primitive reactions of very young children in the presence 
of marbles. 

But in order to know to what consciousness of rules 
these individual schemas correspond it should be re- 
membered that from its tenderest years everything con- 
spires to impress upon the baby the notion of regularity. 
Certain physical events (alternation of day and night, 
sameness of scenery during walks, etc.) are repeated with 
sufficient accuracy to produce an awareness of "law", 


or at any rate to favour tie appearance of motor schemas 
of prevision. The parents, moreover, impose upon the 
baby a certain number of moral obligations, the source 
of further regularities (meals, bed-time, cleanliness, etc.) 
which are completely (and to the child indissociably) 
connected with the external regularities. From its earliest 
months the child is therefore bathed in an atmosphere 
of rules, so that the task of discerning what comes from 
itself in the rites that it respects and what results from 
the ^ pressure of things or the constraint of the social 
environment is one of extreme difficulty. In the content 
of each ritual act it is certainly possible to know what 
has been invented by the child, what discovered in nature, 
and what imposed by the adult. But in the consciousness 
of rules, taken as a formal structure, these differentiations 
are non-existent from the point of view of the subject 
himself. 1 

An analysis of the rites practised by older children, 
however, will allow us to introduce a fundamental dis- 
tinction at this point. On the one hand, certain forms 
of behaviour are, as it were, ritualized by the child himself 
(e.g. not to walk on the lines that separate the paving 
stones from the kerb of the pavement). Now, so long 
as no other factor intervenes, these motor rules never 
give rise to the feeling of obligation proper. (This is true 
even of the example we selected intentionally just now 
that of a simple game which only becomes obligatory 
when it becomes connected later on with a pact, i.e. with 
a social operation, for the pact with oneself is undoubtedly 
a derivative of the pact with others.) On the other 
hand, certain rules it matters not whether they were 
previously invented by the child, imitated, or received 
from outside are at a given moment sanctioned by the 

1 e.g. Heat burns (physical law), it is forbidden to touch the fire 
(moral law) and the child playing about in the kitchen will amuse 
himself by touching every piece of furniture except the stove (individual 
ritual). How can the subject's mind distinguish at first between these 
three types of regularity ? 


environment, i.e. approved of or enjoined. Only in such 
a case as this are rules accompanied by a feeling of 
obligation. Now, although it is always difficult to know 
to what extent an obligatory rale covers up in the mind 
of a child of one or two years a motor ritual, it is at any 
rate obvious that the two things are psychologically dis- 
tinct. And this distinction should be borne in mind when 
we come to the study of the rules of the game. 

The reader will recognize in the way in which we have 
stated the problem the striking thesis of M. Bovet on 
the genesis of the feeling of moral obligation in man's 
conscience : the feeling of obligation only appears when 
the child accepts a command emanating from someone 
whom he respects. All the material analysed in the present 
work, beginning with the facts relating to consciousness 
of the rules of the game, confirm this thesis, which is 
parallel rather than contradictory to Durkheim's doctrine 
of the social genesis of respect and morality. The only 
change we wish to effect in Bovet's theory is to extend it 
and to introduce alongside of the unilateral respect of the 
younger child for the grown-up, the mutual respect that 
is entertained among equals. Consequently, a collective 
rule will appear to us as much the product of the reciprocal 
approbation of two individuals as of the authority of one 
individual over another. 

What then does consciousness of rules amount to during 
our first stage ? In so far as the child has never seen 
anyone else play, we can allow that it is engaged here 
upon purely personal and individual ritual acts. The 
child, enjoying as it does any form of repetition, gives 
itself schemas of action, but there is nothing in this that 
implies an obligatory rule. At the same time, and this is 
where the analysis becomes so difficult, it is obvious that 
by the time a child can speak, even if it has never 
seen marbles before, it is already permeated with rules 
and regulations due to the environment, and this in the 
most varied spheres. It knows that some things are 
allowed and others forbidden. Even in the most modern 


form of training one cannot avoid imposing certain 
obligations with regard to sleeping, eating, and even in 
connection with certain details of no apparent importance 
(not to touch a pile of plates, daddy's desk, etc., etc.). 
It is therefore quite possible that when the child comes 
across marbles for the first time, it is already convinced 
that certain rules apply to these new objects. And this 
is why the origins of consciousness of rules even in so 
restricted a field as that of the game of marbles are 
conditioned by the child's moral life as a whole. 

This becomes clear in the second stage, the most 
interesting for our thesis. This second stage sets in from 
the moment when the child, either through imitation or 
as the result of verbal exchange, begins to want to play 
in conformity with certain rules received from outside. 
What idea does he form of these rules ? This is the point 
that we must now try to establish. 

We made use of three groups of questions for the 
purpose of analysing the consciousness of rules in this 
second stage. Can rules be changed ? Have rules always 
been the same as they are to-day ? How did rules begin ? 
Obviously the first of these questions is the best. It is 
the least verbal of the three. Instead of making the 
child think about a problem that has never occurred to 
him (as do the other two), it confronts the subject with a 
new fact, a rule invented by himself, and it is relatively 
easy to note the child's resulting reactions, however 
clumsy he may be in formulating them. The other 
two questions, on the contrary, incur all the objections 
that can be made against questioning pure and simple 
the possibility of suggestion, of perseveration, etc. We 
are of opinion, nevertheless, that these questions have 
their use, if only as indices of the respect felt for rules 
and as complementary to the first. 

Now, as soon as the second stage begins, i.e. from the 
moment that the child begins to imitate the rules of 
others, no matter how egocentric in practice his play may 
be, he regards the rules of the game as sacred and un- 


touchable ; he refuses to alter these rules and claims that 
any modification, even if accepted by general opinion, 
would be wrong. 

Actually, it is not until about the age of 6 that this 
attitude appears quite clearly and explicitly. Children of 
4-5 seem, therefore, to form an exception and to take 
rules rather casually, a feature which, if judged purely 
externally, recalls the liberalism of older children. In 
reality, we believe that this analogy is superficial, and 
that little children, even when they seem not to be so, 
are always conservative in the matter of rules. If they 
accept innovations that are proposed to them, it is 
because they do not realize that there was any innovation. 

Let us begin by one of the more difficult cases, the 
difficulty being all the greater because the child is very 
young and consequently very much inclined to romance. 

FAL (5) is at the second stage with regard to the 
practice of rules. " Long ago when people were beginning 
to build the town of Neuchitel, did little children play 
at marbles the way you showed me ? Yes. Always that 
way ? Yes, How did you get to know the rules ? 
When I was quite little my brother showed me. My Daddy 
showed my brother. And how did your daddy know? 
My Daddy just knew. No one told him. How did he 
know ? No one showed him ! " " Am I older than your 
Daddy ? No, you're young. My Daddy had been born 
when we came to Neuchdtel. My Daddy was born before 
me. Tell me some people older than your daddy. My 
grand-dad. Did he play marbles ? Yes. Then he played 
before your daddy ? Yes, but not with rules ! [said with 
great conviction]. What do you mean by rules ? . . . 
[Fal does not know this word, which he~ has just heard 
from our lips for the first time. But he realizes that it 
means an essential property of the game of marbles ; 
that is why he asserts so emphatically that his grand-dad 
did not play with rules so as to show how superior his 
daddy is to everyone else in the world.] Was it a long 
time ago when people played for the first time ? Oh, 
yes. How did they find out how to play ? Well, they 
took some marbles, and then they made a square, and then 
they put the marbles inside it . . . etc. [he enumerates the 


rules that he knows]. Was it little children who found 
out or grown-up gentlemen ? Grown-up gentlemen. 
Tell me who was born first, your daddy or your grand- 
dad ? My Daddy was born before my grand-dad. Who 
invented the game of marbles ? My Daddy did. Who 
is the oldest person in Neuchatel ? / dunno. Who do 
you think ? God. Did people know how to play marbles 
before your daddy ? Other gentlemen played [before ? 
at the same time ?]. In the same way as your daddy ? 
Yes. How did they know how to ? They made it up. 
Where is God ? In the sky. Is he older than your daddy ? 
Not so old." " Could one find a new way of playing ? 
I can't play any other way. Try . . . [Fal does not 
move]. Couldn't you put them like this [we place the 
marbles in a circle without a square] ? Oh, yes. Would 
it be fair ? Oh, yes. As fair as the square ? Yes. 
Did your daddy use to play that way or not ? Oh, yes. 
Could one play still other ways ? Oh, yes." We then 
arrange the marbles in the shape of a T, we put them 
on a matchbox, etc. Fal says he has never seen this 
done before, but that it is all quite fair and that you can 
change things as much as you like. Only his daddy 
knows all this ! 

Fal is typical of the cases we were discussing above. 
He is ready to change all the established rules. A circle, 
a T., anything will do just as well as the square. It looks, 
at first, as though Fal were near those older children 
who, as we shall see, no longer believe in the sacred 
character of rules and adopt any convention so long as 
it is received by all. But in reality this is not the case. 
However great a romancer Fal may be, the text of which 
we have quoted the greater part seems to show that he 
has a great respect for rules. He attributes them to his 
father, which amounts to saying that he regards them 
as endowed with divine right. Fal's curious ideas about 
his father's age are worth noting in this connection ; his 
daddy was born before his grand-dad, and is older than 
God ! These remarks, which fully coincide with those 
collected by M. Bovet, 1 would seem to indicate that in 
attributing the rules to his father, Fal makes them more 

1 P. Bovet, The Child's Religion, London, 1930. 


or less contemporaneous with what is for him the beginning 
of the world. Characteristic, too, is the manner in which 
the child conceives this invention of rules on the part 
of his father : this gentleman thought of them without 
having been told or shown anything, but other gentlemen 
may equally have thought of the same thing. This is 
not, in our opinion, mere psittacism. One should be 
careful, of course, not to read into these remarks more 
logic than they contain : they simply mean that rules 
are sacred and unchangeable because they partake of 
paternal authority. But this affective postulate can be 
translated into a sort of infantile theory of invention, and 
of the eternity of essences. To the child who attaches no 
precise meaning to the terms " before " and " after " and 
who measures time in terms of his immediate or deeper 
feelings, to invent means almost the same thing as to 
discover an eternal and pre-existing reality in oneself. Or 
to put it more simply, the child cannot differentiate as 
we do between the activity which consists in inventing 
something new and that which consists in remembering 
the past. (Hence the mixture of romancing and exact 
reproduction which characterizes his stories or his memory.) 
For the child, as for Plato, intellectual creation merges 
into reminiscence, 1 What, then, is the meaning of Fal's 
tolerance with regard to the new laws we suggested to 
him ? Simply this, that confident of the unlimited 
wealth of rules in the game of marbles, he imagines, as 
soon as he is in possession of a new rule, that he has 
merely rediscovered a rule that was already in existence. 

In order to understand the attitude of the children of 
the early part of the second stage they all answer more 
or less like Fal we must remember that up till the age 
of 6-7 the child has great difficulty in knowing what 
comes from himself and what from others in his own 
fund of knowledge. This comes primarily from his 

1 Cf. C.W., p. 52, the case of Kauf ( 8 ; 8) : this child believes that 
the stones she tells were written in her brain by God. " Before I was 
born, he put them there." 


'difficulty in retrospection (see /. R., Chap. IV, i), and 
secondly from the lack of organization in memory itself. 
In this way the child is led to think that he has always 
known something which in fact he has only just learned. 
We have often had the experience of telling a child 
something which immediately afterwards he will imagine 
himself to have known for months. This indifference to 
distinctions of before and after, old and new, explains 
the inability of which we spoke just now to differentiate 
between invention and reminiscence. The child very often 
feels that what he makes up, even on the spur of the 
moment, expresses, in some way, an eternal truth. This 
being so, one cannot say that very young children have 
no respect for rules because they allow these to be changed ; 
innovations are not real innovations to them. 

Added to this there is a curious attitude which appears 
throughout the whole of the egocentric stage, and which 
may be compared to the mental states characteristic of 
inspiration. The child more or less pleases himself in his 
application of the rules. At the same time, Fal and others 
like him will allow any sort of change in the established 
usage. And yet they one and all insist upon the point 
that rules have always been the same as they are at 
present, and that they are due to adult authority, 
particularly the authority of the father. Is this con- 
tradictory ? It is so only in appearance. If we call to 
mind the peculiar mentality of children of this age, for 
whom society is not so much a successful cooperation 
between equals as a feeling of continuous communion 
between the ego and the Word of the Elder or Adult, 
then the contradiction ceases. Just as the mystic can 
no longer dissociate his own wishes from the will of his 
God, so the little child cannot differentiate between the 
impulses of his personal fancy and the rules imposed on 
him from above. 

Let us now pass on to the typical cases of this stage, 
i.e. to children who out of respect to rules are hostile to 
any innovation whatsoever. 


We must begin by quoting a child of 5| years, LEH, 
whose reaction was among the most spontaneous that 
we had occasion to note. Leh was telling us about the 
rules of the game before we had questioned him about 
consciousness of rules. He had just begun to speak and 
was showing us how to play from the coche (which was 
about the only thing in the game that he knew) when 
the following dualogue took place. We asked Leh quite 
simply if everyone played from the coche or whether one 
could not (as is actually done) put the older ones at the 
coche and let the little ones play closer up. "No, 
answered Leh, that wouldn't be fair. Why not ? Because 
God would make the little boy's shot not reach the marbles 
and the big boy's shot would reach them." In other words, 
divine justice is opposed to any change in the rules of 
marbles, and if one player, even a very young one were 
favoured in any way, God Himself would prevent him 
from reaching the square. 

PHA (sJ) : " Do people always play like that ? Yes, 
always like that. Why ? 'Cos you couldn't play any other 
way ^ Couldn't you play like this [we arrange the marbles 
in a circle, then in a triangle] ? Yes, but the others wouldn't 
wan i 20. Why ? 'Cos squares is better. Why better ? 
. . ." We are less successful, however, with regard 
to the origins of the game : " Did your daddy play at 
marbles before you were born ? No, never, because I 
wasn't there yet! But he was a child like you before 
you were born, / was there already when he was like 
me. He was bigger." " When did people begin to play 
marbles ? When the others began, I began too." It would 
be impossible to outdo Pha in placing oneself at the 
centre of the universe, in time as well as in space ! And 
yet Pha feels very strongly that rules stand above him : 
they cannot be changed. 

GEO (6) tells us that the game of marbles began \yith 
" people, with the Gentlemen of the Commune [the Town 
Council whom he has probably heard mentioned in 
connection with road-mending and the police]. How was 
that ? It came into the gentlemen's heads and they made 
some marbles. How did they know how to play ? In 
their head. They taught people. Daddies show little boys 
how to. Can one play differently from how you showed 
me ? Can you change the game ? I think you can, but 
I don't know how [Geo is alluding here to the variants 
already in existence]. Anyhow ? No there are no games 


you play anyhow. Why ? Because God didn't leach 
them [the Town Council]. Try and change the game. 
[Geo then invents an arrangement which he regards as 
quite new and which consists in making a big square 
with three rows of three marbles each], Is that one 
fair, like the other one ? No, becatise there are only three 
lines of three. Could people always play that way and 
stop playing the old way ? Yes, M'sieu. How did you 
find this game ? In my head. -Can we say, then, that 
the other games don't count and this is the one people 
must take ? Yes, M'sieur. There's others too that the 
Gentlemen of the Commune know. Do they know this 
one that you have made up ? Yes [/]. But it was you 
who found it out. Did you find that game in your 
head ? Yes. How ? All of a sudden. God told it to 
me. You know, I have spoken to the gentlemen of the 
Commune, and I don't think they know your new game. 
Oh ! [Geo is very much taken aback]. But I know some 
children who don't know how to play yet. Which game 
shall I teach them, yours, or the other one ? The one 
of the Gentlemen of the Commune. Why ? Because it is 
prettier." " Later on when you are a big man and have 
got moustaches perhaps there won't be many children 
left who play the game of the Gentlemen of the Commune. 
But there may be lots of boys who play at your game. 
Then which game will be fairest, yours, which will be 
played most, or the game of the Gentlemen of the Com- 
mune, which will be nearly forgotten ? The game of the 
Gentlemen of the Commune" 

The case of Geo comes as a beautiful confirmation 
of what we said in connection with Fal, viz. that for 
little children inventing a game comes to the same 
thing as finding in one's head a game that has already 
been anticipated and classified by the most competent 
authorities. Geo attributes the game he has invented to 
divine inspiration, and supposes it to be already known 
to the "Gentlemen of the Commune." As soon as we 
undeceive him he undervalues his own invention and 
refuses to regard it as right even if ratified by general 

MAR (6), whose behaviour in the practice of rules we 
have already examined in 3, declares that in the time 


of his daddy and of Jesus, people played as they do now. 
He refuses to invent a new game. " I've never invented 
games" We then suggest a new game which consists of 
putting marbles on a box and making them fall off by 
hitting the box : " Can one play like this IYes [He does 
so, and seems to enjoy it]. Could this game ever become 
a fair game ? No, because it's not the same. 1 ' Another 
attempt calls forth the same reaction. 

STOR (7) tells us that children played at marbles before 
Noah's ark : " How did they play ? Like we played. 
How did it begin ? They bought some marbles. But how 
did they learn ? His daddy taught them" Stor invents 
a new game in the shape of a triangle. He admits that 
his friends would be glad to play at it, " but not all of them. 
Not the big ones, the quite big ones. Why ? Because it 
isn't a game for the big ones. Is it as fair a game as ^ the 
one you showed me ? No. Why ? Because it isn't a 
square. And if everyone played that way, even the big 
ones, would it be fair ? No. Why not ? Because it isn't 
a square" 

With regard to the practical application of rules all 
these children therefore belong to the stage of egocentrism. 
The result is clearly paradoxical. Here are children 
playing more or less as they choose ; they are influenced, 
it is true, by a few examples that have been set before 
them and observe roughly the general schema of the 
game ; but they do so without troubling to obey in 
detail the rules they know or could know with a little 
attention, and without attributing the least importance 
to the most serious infringements of which they may be 
guilty. Besides all this, each child plays for himself, he 
pays no attention to his neighbour, does not seek to 
control him and is not controlled by him, does not even 
try to beat him " to win " simply means to succeed in 
hitting the marbles one has aimed at. And yet these 
same children harbour an almost mystical respect for 
rules : rules are eternal, due to the authority of parents, 
of the Gentlemen of the Commune, and even of an 
almighty God. It is forbidden to change them, and even 
if the whole of general opinion supported such a change, 
general opinion would be in the wrong : the unanimous 


consent of all the children wauld be powerless against 
the truth of Tradition. As to any apparent changes, 
these are only complementary additions to the initial 
Revelation : thus Geo (the most primitive of the above 
cases, and therefore nearest to those represented by Fal 
and so confirming what we said about the latter) believes 
the rule invented by Mm to be directly due to a divine 
inspiration analogous to the inspiration of which the 
Gentlemen of the Commune were the first recipients. 

In reality, however, this paradox is general in child 
behaviour and constitutes, as we shall show towards the 
end of the book, the most significant feature of the 
morality belonging to the egocentric stage. Childish 
egocentrism, far from being asocial, always goes hand 
in hand with adult constraint. It is presocial only in 
relation to cooperation. In all spheres, two types of 
social relations must be distinguished ; constraint and 
cooperation. The first implies an element of unilateral 
respect, of authority and prestige ; the second is simply 
the intercourse between two individuals on an equal 
footing. Now egocentrism is contradictory only to 
cooperation, for the latter alone is really able to socialize 
the individual. Constraint, on the other hand, is always 
the ally of childish egocentrism. Indeed it is because 
the child cannot establish a genuinely mutual contact 
with the adult that he remains shut up in his own ego. 
The child is, on the one hand, too apt to have the illusion 
of agreement where actually he is only following his own 
fantasy ; the adult, on the other, takes advantage of his 
situation instead of seeking equality. With regard to 
moral rules, the child submits more or less completely 
in intention to the rules laid down for him, but these, 
remaining, as it were, external to the subject's conscience, 
do not really transform his conduct. This is why the 
child looks upon rules as sacred though he does not really 
put them into practice. 

As far as the game of marbles is concerned, there is 
therefore no contradiction between the egocentric practice 


of games and the mystical respect entertained for rales. 
This respect is the mark of a mentality fashioned, not by 
free cooperation between equals, but by adult constraint. 
When the child imitates the rules practised by his older 
companions he feels that he is submitting to an unalterable 
law, due, therefore, to his parents themselves. Thus 
the pressure exercised by older on younger children is 
assimilated here, as so often, to adult pressure. This 
action of the older children is still constraint, for co- 
operation can only arise between equals. Nor does the 
submission of the younger children to the rules of the 
older ones lead to any sort of cooperation in action ; 
it simply produces a sort of mysticism, a diffused feeling 
of collective participation, which, as in the case of many 
mystics, fits in perfectly well with egocentrism. For 
we shall see eventually that cooperation between equals 
not only brings about a gradual change in the child's 
practical attitude, but that it also does away with the 
mystical feeling towards authority. 

In the meantime let us examine the subjects of the 
final period of the present stage. We found only three 
stages with regard to consciousness of rules, whereas 
there seemed to be four with regard to the practice of 
the game. In other words, the cooperation that sets in 
from the age of 7-8 is not sufficient at first to repress 
the mystical attitude to authority, and the last part of 
the present stage (in the consciousness of rules) really 
coincides with the first half of the cooperative stage (in 
the practice of the game). 

BEN (10 yrs.), whose answers we have given with regard 
to the practice of rules (third stage) is still at the second 
stage from the point of view that is occupying us just 
now : " Can one invent new rules ? Some boys do, so 
as to win more marbles, but it doesn't always come off. 
One chap [quite recently, in his class] thought of saymg 
' Deux Empans ' (two spans) so as to get nearer [actually 
this is a rule already known to the older boys]. It didn't 
come off. And with the little ones ? Yes, it came off 
all right with them. Invent a rule. / couldn't invent one 


straight away like that. Yes you could. I can see that 
you are cleverer than you make yourself out to be. 
Well, let's say that you're not caught when you are in the 
square. Good. Would that come off with the others ? 
Oh, yes, they'd like to do that. Then people could play 
that way ? Oh, no, because it would be cheating. But 
all your pals would like to, wouldn't they ? Yes, they 
all would. Then why would it be cheating ? Because 
I invented it : it isn't a rule ! It's a wrong rule because 
it's outside of the rules. A fair rule is one that is in the 
game. How does one know If it is fair ? The good players 
know it. And suppose the good players wanted to play 
with your rule ? It wouldn't work. Besides they would 
say it was cheating. And if they all said that the rale 
was right, would it work ? Oh, yes, it would. . . . But 
it's a wrong rule ! But if they all said it was right how 
would anyone know that it was wrong ? Because when 
vou are in the square it's like a garden with a fence, you're 
"shut in [so that if the shooter stays inside the square, 
you are ' dished ']. And suppose we draw a square like 
this [we draw a square with a break in one of the sides 
like a fence broken by a door] ? Some boys do that. But 
it isn't fair. It's just for fun for passing the time. Why ? 
Because the square ought to be closed. But if some boys 
do it, is it fair or not ? Ifs both fair and not fair. Why 
is it fair ? It is fair for waiting [for fun]. And why is 
it not fair ? Because the square ought to be closed. 
When you are big, suppose everyone plays that way, 
will it be right or not ? It will be right then because there 
will be new children who will learn the rule. And for 
y OU ? ft w m be wrong. And what will it be ' really 
and truly ' ? It will really be wrong." Later on, however, 
Ben admits that his father and grandfather played 
differently from him, and that rules can therefore be 
changed by children. But this does not prevent ^hira 
from sticking to the view that rules contain an intrinsic 
truth which is independent of usage. 

Borderline cases like these are particularly interesting. 
Ben stands midway between the second and third stages. 
On the one hand, he has already learned, thanks to 
cooperation, the existence of possible variations in the 
use of rules, and he knows, therefore, that the actual 
rules are recent and have been made by children. But 


on the other hand, he believes in the absolute and intrinsic 
truth- of rules. Does cooperation, then, impose upon 
this child a mystical attitude to law similar to the respect 
felt by little children for the commands given them by 
adults ? Or is Ben's respect for the rales of the game 
inherited from the constraint that has not yet been 
eliminated by cooperation ? The sequel will show that 
the latter interpretation is the right one. Older children 
cease to believe in the intrinsic value of rules, and they 
do so in the exact measure that they learn to put them 
into practice. Ben's attitude should therefote be regarded 
as a survival of the features due to constraint. 

Generally speaking, it is a perfectly normal thing that 
in its beginnings cooperation on the plane of action 
should not immediately abolish the mental states created 
on the plane of thought by the complexus : ego- 
centricity and constraint. Thought always lags behind 
action and cooperation has to be practised for a very 
long time before its consequences can be brought fully 
to light by reflective thought. This is a fresh example 
of the law of prise de conscience or conscious realization 
formulated by Clapar&de 1 and of the time-lag 2 or " shift- 
ing " which we have observed in so many other spheres (see 
J. R. t Chap. V, 2 and C.C., 2nd part) . A phenomenon such 
as this is, moreover, well fitted to simplify the problem 
of egocentrism in general since it explains why intellectual 
egocentrism is so much more stubborn than egocentrism 
in action. 

After the age of 10 on the average, i.e. from the second 
half of the cooperative stage and during the whole of 

1 This term (Claparede's prise de conscience) simply means "coming 
into consciousness," and has nothing to do with intellectual formula- 
tion. [Trans.]. 

2 This is the term that has been selected by the author ior the French 
decalage, a somewhat more complex notion which in previous volumes, 
cf. L.T., p. 208, ff., has been rendered as a process of "shifting," 



the stage when the rules are codified, consciousness of 
rules undergoes a complete transformation. Autonomy 
follows upon heteronomy : the rule of a game appears 
to the child no longer as an external law, sacred in so 
far as it has been laid down by adults ; but as the 
outcome of a free decision and worthy of respect in the 
measure that it has enlisted mutual consent. 

This change can be seen by three concordant symptoms. 
In the first place, the child allows a change in the rules 
so long as it enlists the votes of all. Anything is possible, 
so long as, and to the extent that you undertake to 
respect the new decisions. Thus democracy follows on 
theocracy and gerontocracy : there are no more crimes 
of opinion, but only breaches in procedure. All opinions 
are tolerated so long as their protagonists urge their 
acceptance by legal methods. Of course some opinions 
are more reasonable than others. Among the new rules 
that may be proposed, there are innovations worthy of 
acceptance because they will add to the interest of the 
game (pleasure in risks, art for art's sake, etc.). And 
there are new rules that are worthless because they give 
precedence to easy winning as against work and skill. 
But the child counts on the agreement among the players 
to eliminate these immoral innovations. He no longer 
relies, as do the little ones, upon an all-wise tradition. 
He no longer thinks that everything has been arranged 
for the best in the past and that the only way of avoiding 
trouble is by religiously respecting the established order. 
He believes in the value of experiment in so far as it is 
sanctioned by collective opinion. 

In the second place, the child ceases ipso facto to look 
upon rules as eternal and as having been handed down 
unchanged from one generation to another. Thirdly and 
finally, his ideas on the origin of the rules and of the 
game do not differ from ours : originally, marbles must 
simply have been rounded pebbles which children threw 
about to amuse themselves, and rules, far from having 
been imposed as such by adults, must have become 


gradually fixed on the initiative of the children them- 

Here are examples : 

Ross (n) belongs to the third stage in regard to the 
practice of rules. He claims that he often invents new 
rales with Ms playmates : " We make them [up] some- 
times We go up to 200. We play about and then hit each 
other', and then he says to me : ' If you & up to 100 I'U 
give you a marble.' Is this new rule fair like the old 
ones or not ? Perhaps it isn't quite fair, because ^t ^snt 
very ' hard to take four marbles that way f If everyone 
does it will it be a real rule, or not IIf they do it often, 
it will become a real rule. Did your father play the way 
you showed me, or differently ? Oh, I don't know. It 
may have been a different game. It changes. It still changes 
quite often. Have people been playing for long ? A t 
least fifty years. Did people play marbles in the days 
of the 'Old Swiss ' IOh, I don't think so. Row did it 

begin ? Some boys took some motor balls (ball bearings) 

and then they played. And after that there were marbles 
in shops. Why are there rules in the game of marbles ? 
So as not to be always quarrelling you must have rules, and 
then play properly How did these rules begin ISome 
boys came to an agreement amongst themselves and made 
them. Could you invent a new rule ? Perhaps ... [he 
thinks] you put three marbles together and you drop another 
from above on to the middle one. Could one play that way ? 
Oh, yes. IB that a fair rule like the others IThe 
chaps might say it wasn't very fair because it's luck. To 
be a good rule, it has to be skill. But if everyone played 
that way, would it be a fair rule or not ? Oh 9 yes, you 
could play just as well with that rule as with the others" 

MALE (12) belongs to the fourth stage in the practice 
of rules : " Does everyone play the way you showed 
me 7 Yes. And did they play like that long ago ? 

N0. Why not ? They used different words. And how 

about the rules ? They didn't use them either, because 
my father told me he didn't play that way. But long ago 
did people play with the same rules 7 Not quite the 
same. How about the rule not hitting for one ? I think 
that must have come later.Did they play marbles when 
your grandfather was little ? Yes. Like they do now ? 

Oh, no, different kinds of games. And at the time 

of the battle of Morat ? No, I don't think they played 


. How do you think the game of marbles began ? 
At first, children looked for round pebbles. And the rules ? 
/ expect they played from the cache. Later on, boys 
wanted to play differently and they invented other rules. 
And how did the coche begin ? I expect they had fun 
hitting the pebbles. And then they invented the coche. 
Could one change the rales IYes. Could you ? Yes t 
I could make up another game. We were playing ^ at home 
one evening and we found out a new one [he shows it to us]. 
Are these new rules as fair as the others ? Yes. 
Which is the fairest, the game you showed me first or 
the one you invented ? Both the same. If you show 
this new game to the little ones what will they do ? 
Perhaps they will play at it. And if they forget the 
square game and only play this one, which will be the 
true game, the new one that will be better known, or the 
old one ? The best known one will be the fairest." 

GROS (13 yrs. at the fourth stage in the practice of the 
rules) has shown us the rules as we saw above. " Did 
your father play that way when he was little ? No, 
they had other rules. They didn't play with a square. 
And did the other boys of your father's time play with 
a square ? There must have been one who knew, since 
we know it now. And how did that one know about 
the square ? They thought they would see if it was nicer 
than the other game. How old was the boy who invented 
the square ? J expect thirteen [his own age]. Did the 
children of the Swiss who lived at the time of the battle 
of Morat play at marbles ? They may have played with 
a hole, and then later on with a square. And in the time 
of David de Purry [a periwigged gentleman whose 
statue on one of the public squares of NeucMtel is known 
to all] ? / expect they had a bit of a lark too ! Have 
rules changed since the square was invented? There 
may have been little changes. And do the rules still change ? 
No. You always play the same way. Are you aEowed 
to change the rules at all IOh, yes. Some want to, and 
some don't. ' If the boys play that way (changing some- 
thing) you have to play like they do. Do you think you 
could invent a new rule ? Oh, yes . . . [he thinks] ; 
you could play with your feet. Would it be fair? / 
don't know. It's just my idea. And if you showed it to 
the others would it work fIt would work all right. Some 
-other boys would want to try. Some wouldn't, by Jove! 
They would stick to the old rules. They'd think they'd have 


less of a chance with this new game. And if everyone 
played your way ? Then it would be a rule like the others. 
Which is the fairest now, yours or the old one ? The 
old one. Why ? Because they cant cheat. (Note this 
excellent justification of rules : the old rule is better 
than the innovation, not yet sanctioned by usage, because 
only the old rule has the force of a law and can thus 
prevent cheating.) And if nearly everyone played with 
their feet, then which would be fairest ? // nearly 
everyone played with their feet, then that would be the 
fairest." Finally we ask Gros, " Suppose there are two 
games, an easy one where you win often, and a difficult 
one where you win seldom, which would you like best ? 
The most difficult. You end by winning that way" 

VUA (13), whose answers about the practice of rules 
we have already examined (4th stage) tells us that his 
father and his grandfather played differently from him. 
" In the days of the ' Three Swiss ' did boys play at 
marbles ? No. They had to work at home. They played 
other games. Did they play marbles in the days of the 
battle of Morat ? Perhaps, after the war. Who invented 
this game ? Some kids. They saw their parents playing 
at bowls, and they thought they might do the same thing. 
Could other rules be invented ? Yes [he shows us one 
he has invented and which he calls ' the line * because 
the marbles are arranged in a row and not in a square]. 
Which is the real game, yours or the square ? The 
square, because it is the one that is always used. Which 
do you like best, an easy game or a difficult one ? 
The more difficult, because it is more interesting. The 
' Troyat ' (a game that consists in heaping the balls into 
piles) is not quite the real game. Some boys invented it. 
They wanted to win all the marbles." On this point Vua 
seems to be answering like a child of the preceding stage 
who will invoke the " real game " that conforms to 
tradition as against contemporary innovations. But Vua 
seems to us rather to be contrasting a demagogic pro- 
cedure (the " Troyat/' which by allowing too great a part 
to chance gives rise to illicit and immoral gains) with 
practices that are in keeping with the spirit of the game, 
whether they are ancient, like the square, or recent like 
his own game. The proof of this would seem to lie in 
the following remarks relating to his own playing : "Is 
the game you invented as fair as the square, or less 
fair ? It is just as fair because the marbles are far apart 


(therefore the game is difficult). If in a few years 1 time 
everyone played your line game and only one or two 
boys played the square game, which would be the fairest, 
the line or the square ? The line would be fairest!' 

BLAS (12, 4th stage in the practice of rules) thinks 
that the game of marbles must have begun round about 
1500 at the time of the Reformation. " Children invented 
the game. They made little balls with earth and water and 
then they amused themselves by rolling them about. They 
found it was rather fun making them hit, and then they had 
the idea of inventing a game, and they said that when you 
hit anyone else's marble with your own you could have the 
marble you hit. After that I expect they invented the square, 
so that you should have to make the marbles go out of the 
square. They invented the line, so that all the marbles 
should be at the same distance. They only invented it later. 
When cement was discovered, marbles were made like they 
are to-day. The marbles of earth weren't strong enough, so 
the children asked the manufacturers to make some in 
cement." We ask Bias to make up a new rule, and this is 
what he thinks of. First there must be a competition, and 
whoever makes his marbles go furthest can play first. But 
the rule seems " bad because you'd have to run too far back to 
fetch the marbles." He then thinks of another which consists 
in playing in two squares one inside the other. " Would 
everyone want to play that way ? Those who invented it 
would. Later on, if your game is played just as much as 
the square, which will be the fairest ? Both the same" 

The psychological and educational interest of all this 
stands out very clearly. We are now definitely in the 
presence of a social reality that has rational and moral 
organization and is yet peculiar to childhood. Also we 
can actually put our finger upon the conjunction of 
cooperation and autonomy, which follows upon the 
conjunction of egocentrism and constraint. 

Up to the present, rules have been imposed upon the 
younger children by the older ones. As such they had 
been assimilated by the former to the commands given 
by adults. They therefore appeared to the child as 
sacred and untouchable, the guarantee of their truth 
being precisely this immutability. Actually this con- 
formity, like all conformity, remained external to the 


individual. In appearance docile, in his own eyes sub- 
missive and constantly imbued as it were with the spirit 
of the Elders or the Gods, the child could in actual fact 
achieve little more than a simulation of sociality, to say 
nothing of morality. External constraint does not 
destroy egocentrism. It covers and conceals when it 
does not actually strengthen it. 

But from henceforward a rule is conceived as the free 
pronouncement of the actual individual minds them- 
selves. It is no longer external and coercive : it can 
be modified and adapted to the tendencies of the group. 
It constitutes no revealed truth whose sacred character 
derives from its divine origin and historical permanence ; 
it is something that is built up progressively and auto- 
nomously. But does this not make it cease to be a real 
rule ? Is it perhaps not a mark of decadence rather 
than of progress in relation to the earlier stage ? That 
is the problem. The facts, however, seem definitely to 
authorize the opposite conclusion : it is from the moment 
that it replaces the rule of constraint that the rule of 
cooperation becomes an effective moral law. 

In the first place, one is struck by the synchronism 
between the appearance of this new type of consciousness 
of rules and a genuine observation of the rules. This 
third stage of rule consciousness appears towards the age 
of lo-n. And it is at this same age that the simple 
cooperation characteristic of the third stage in the 
practice of rules begins to be complicated by a desire 
for codification and complete application of the law. 
The two phenomena are therefore related to each other. 
But is it the consciousness of autonomy that leads to the 
practical respect for the law, or does this respect for the 
law lead to the feeling of autonomy ? These are simply 
two aspects of the same reality : when a rule ceases to 
be external to children and depends only on their free 
collective will, it becomes incorporated in the mind of 
each, and individual obedience is henceforth purely 
spontaneous. True, the difficulty reappears each time 


that the child, while still remaining faithful to a rule that 
favours him, is tempted to slur over some article of the 
law or some point of procedure that favours his opponent. 
But the peculiar function of cooperation is to lead the 
child to the practice of reciprocity, hence of moral univer- 
sality and generosity in Ms relations with his playmates. 

This last point introduces us to yet another sign of 
the bond between autonomy and true respect for the 
law. By modifying rules, i.e. by becoming a sovereign 
and legislator in the democracy which towards the age 
of lo-n follows upon the earlier gerontocracy, the child 
takes cognizance of the raison d'etre of laws. A rule 
becomes the necessary condition for agreement, " So as 
not to be always quarrelling,'* says Ross, ''you must have 
rules and then play properly [ = stick to them]/' The 
fairest rule, Gros maintains, is that which unites the 
opinion of the players, " because [then] they can't cheat. 1 ' 

Thirdly, what shows most clearly that the autonomy 
achieved during this stage leads more surely to respect 
for rules than the heteronomy of the preceding stage 
is the truly political and democratic way in which children 
of 12-13 distinguish lawless whims from constitutional 
innovation. Everything is allowed, every individual 
proposition is, by rights, worthy of attention. There are 
no more breaches of opinion, in the sense that to desire 
to change the laws is no longer to sin against them. Only 
and each of our subjects was perfectly clear on this 
point no one has the right to introduce an innovation 
except by legal channels, i.e. by previously persuading 
the other players and by submitting in advance to the 
verdict of the majority. There may therefore be breaches 
but they are of procedure only : procedure alone is 
obligatory, opinions can always be subjected to discussion. 
Thus Gros tells us that if a change is proposed " Some 
want to and some don't. If boys play that way [allow an 
alteration] you have to play like they do." As Vua said 
in connection with the practice of rules (4) " sometimes 
people play differently. Then you ask each other what you 

want to do. . . . We scrap for a bit and then we fix things 

In short, law now emanates from the sovereign people 
and no longer from the tradition laid down by the Elders. 
And correiatively with this change, the respective values 
attaching to custom and the rights of reason come to be 
practically reversed. 

In the past, custom had always prevailed over rights. 
Only, as in all cases where a human being is enslaved to 
a custom that is not part of his inner life, the child 
regarded this Custom imposed by his elders as a sort of 
Decalogue revealed by divine beings (i.e. adults, including 
God, who is, according to Fal, the oldest gentleman in 
Neuchatel after his own father). With the result that, 
in the eyes of a little child, no alteration of usage will 
dispense the individual from remaining faithful to the 
eternal law. Even if people forget the square game, 
says Ben, and adopt another, this new game " will really 
be wrong." The child therefore distinguishes between a 
rule that is true in itself and mere custom, present or 
future. And yet he is all the time enslaved to custom 
and not to any juridico-raoral reason or reality distinct 
from this custom and superior to it. Nor indeed is this 
way of thinking very different from that of many con- 
servative adults who delude themselves into thinking 
that they are assisting the triumph of eternal reason 
over present fashion, when they are really the slaves of 
past custom at the expense of the permanent laws of 
rational cooperation. 

But from now on, by the mere fact of tying himself down 
to certain rules of discussion and collaboration, and thus 
cooperating with his neighbours in full reciprocity (without 
any false respect for tradition nor for the will of any one 
individual) the child will be enabled to dissociate custom 
from the rational ideal. For it is of the essence of coopera- 
tion as opposed to social constraint that, side by side with 
the body of provisional opinion which exists in fact, it also 
allows for an ideal of what is right functionally implied in the 


very mechanism of discussion and reciprocity. The 
constraint of tradition imposes opinions or customs, and 
there is an end of it. Cooperation imposes nothing 
except the actual methods of intellectual or moral inter- 
change (Baldwin's * synnomic as opposed to his syndoxic). 
Consequently we must distinguish alongside of the actual 
agreement that exists between minds, an ideal agreement 
defined by the more and more intensive application of 
the processes of mental interchange. 2 As far as our 
children are concerned, this simply means that in addition 
to the rules agreed upon in a given group and at a given 
moment (constituted morality or rights in the sense in 
which M. Lalande speaks of " raison constitute " 3 ) the 
child has in mind a sort of ideal or spirit of the game 
which cannot be formulated in terms of rules (constitutive 
morality or rights in the sense of " raison constituante "). 
For if there is to be any reciprocity between players in 
the application of established rules or in the elaboration 
of new rules, everything must be eliminated that would 
compromise this reciprocity (inequalities due to chance, 
excessive individual differences in skill or muscular 
power, etc.). Thus usages are gradually purified in virtue 
of an ideal that is superior to custom since it arises from 
the very functioning of cooperation. 

This is why, when innovations are proposed to the 
child, he regards them as fair or unfair not only according 
as they are likely or not to rally the majority of players 
in their favour, but also according as they are in keeping 
with that spirit of the game itself, which is nothing more 
or less than the spirit of reciprocity. Ross tells us, for 
instance, concerning his own proposition, " Perhaps it 
isn't quite fair, because it isn't very hard to take four marbles 
that way/' and again, " The chaps might say it wasn't 
very fair because it's luck. To be a good rule, it has to be 

1 J. M. Baldwin, Genetic Theory of Reality. 

2 See our article, " Logique gene'tique et sociologie," Revue Philo- 
sophique, 1928. 

8 Lalande, A., " Raison constituante et raison constitute/' Revue 
dfs Cours et des Conference. 


skill." The Troyat, Vua informs us, is less fair than the 
square (though equally widespread and equally well 
known to former generations), because it was invented 
" to win all the marbles." In this way, Vua draws a 
distinction between demagogy and a sane democracy. 
In the same way, Gros and Vua prefer difficult games 
because they are more " interesting " : cleverness and 
skill now matter more than winning. Art for art's sake 
is far more disinterested than playing for gain. 

In a word, as soon as we have cooperation, the rational 
notions of the just and the unjust become regulative of 
custom, because they are implied in the actual functioning 
of social life among equals a point which will be developed 
in the third chapter of this book. During the preceding 
stages, on the contrary, custom overbore the issue of 
right, precisely in so far as it was deified and remained 
external to the minds of individuals. 

Let us now see what sort of philosophy of history the 
child will adopt in consequence of having discovered 
democracy. It is very interesting, in this connection, to 
note the following synchronism. The moment a child 
decides that rules can be changed, he ceases to believe 
in their endless past and in their adult origin. In other 
words, he regards rules as having constantly changed 
and as having been invented and modified by children 
themselves. External events may of course play a certain 
part in bringing this about. Sooner or later, for example, 
the child may learn from his father that the game was 
different for previous generations. But so unmistakable 
is the correlation (on the average, of course) between the 
appearance of this new type of consciousness of rules 
and the disappearance of the belief in the adult origin of 
the game that the connection must be founded on reality. 
Is it, then, the loss of belief in the divine or adult origin 
of rules that allows the child to think of innovations, 
or is it the consciousness of autonomy that dispels the 
myth of revelation ? 

Only someone completely ignorant of the character 


of childish beliefs could imagine that a change in the 
child's ideas about the origin of rules could be of a nature 
to exercise so profound an influence on his social conduct. 
On the contrary, here as in so many cases, belief merely 
reflects behaviour. There can be no doubt that children 
very rarely reflect upon the original institution of -the, 
game of marbles. There are even very strong reasons 
for assuming that as far as the children we examined 
are concerned such a problem never even entered their 
heads until the day when a psychologist had the ridiculous 
idea of asking them how marbles were played in the days 
of the Old Swiss and of the Old Testament. Even if the 
question of the origin of rules did pass through the minds 
of some of these children during the spontaneous inter- 
rogatories that so often deal with rules in general (L. T., 
Chap. V, 5 and 10) the answer which the child would 
give himself would probably be found without very much 
reflection. In most cases the questions we asked were 
entirely new to the subject, and the answers were dictated 
by the feelings which the game had aroused in them in 
varying intensity. Thus, when the little ones tell us that 
rules have an adult origin and have never changed, one 
should beware of taking this as the expression of a 
systematic belief ; all they mean is that the laws of the 
game must be left alone. And when, conversely, the 
older ones tell us that rules have varied and were invented 
by children, this belief is perhaps more thought out since 
it is held by more developed subjects, but it is still 
only valuable as an indication : the child simply means 
that he is free to make the law. 

We may well ask ourselves, then, whether it is legiti- 
mate to question the child about such very verbal beliefs, 
since these beliefs do not correspond to thought properly 
so called, and since the child's true thought lies much 
deeper, somewhere below the level of formulation. But 
in our opinion these beliefs have their interest because the 
same phenomena reappear in adult mental life and because 
the psychological facts lead by a series of intermediate 


steps to metaphysical systems themselves. What Pareto, 1 
basing his relatively simple conclusions on such a wealth 
of erudition, has called " derivations " are really present 
in germ in our children's remarks about the origin of 
games. These remarks have no intellectual value, but 
they contain a very resistant, affective and social element 
the "residuum" to quote Pareto again. To the 
residuum peculiar to the conforming attitude of the 
little ones correspond the derivations " divine or adult 
origin " and " permanence in history ". To the residuum 
peculiar to the more democratic attitude of- the older 
children correspond the derivations " natural (childish) 
origin " and " progress ". 

One more fundamental question must still be discussed. 
How is it that democratic practice is so developed in the 
games of marbles played by boys of n to 13, whereas it 
is still so unfamiliar to the adult in many spheres of life ? 
Of course it is easier to agree upon some subjects than on 
others, and feeling will not run so high on the subject 
of the rales of the " Square " as in an argument about 
the laws of property or the legitimacy of war. But 
apart from these questions (and after all, is it so obvious 
that social questions are -more important, to us than 
are the rules of a game to the child of 12 j>) there are 
others of greater psychological and sociological interest. 
For it must not be forgotten that the game of 
marbles is dropped towards the age of 14-15 at the 
latest. With regard to this game, therefore, children 
of 11-13 have no seniors. The following circumstance 
is important. Since they no longer have to endure the 
pressure of play-mates who impose their views by virtue 
of their prestige, the children whose reactions we have 
been studying are clearly able to become conscious of 
their autonomy much sooner than if the game of 
marbles lasted till the age of 18. In the same way, 
most of the phenomena which characterize adult societies 
would be quite other than they are if the average length 

i Tvaitt de Sodologie gln&rale. 


of human life were appreciably different from what it 
is. Sociologists have tended to overlook this fact, 
though Auguste Comte pointed out that the pressure 
of one generation upon the other was the most important 
phenomenon of social life. 

We shall have occasion to see, moreover, that towards 
the age of n the consciousness of autonomy appears in 
a large number of different spheres. Whether this is the 
repercussion of collective games on the whole moral life 
of the child is a question which will be taken up 

drawing any general conclusion from the facts set out 
above, it may be useful to see whether they are peculiar 
to the game of marbles as played by boys or whether 
similar examples cannot be found in different fields. 
For this purpose we studied, with the same method, but 
questioning only girls, a much simpler game than the 
game of marbles. 

The most superficial observation is sufficient to show 
that in the main the legal sense is far less developed in 
little girls than in boys. We did not succeed in finding 
a single collective game played by girls in which there 
were as many rules and, above all, as fine and consistent 
an organization and codification of these rules as in the 
game of marbles examined above. A significant example 
in this connection is the game of " Marelle " (Engl., 
Hop-scotch) (also called "la Semaine " or " le del") 
which consists in hopping on one leg and kicking a stone 
through various sections drawn on the ground representing 
the days of the week or anything else one likes. The 
few rules embodied in this game (not to put the other 
foot down, to make the pebble go into the right square 
with one kick, not to let the pebble stop on a boundary 
line, permission to rest in a special section called Heaven, 
etc,) show well enough how possible it would have been 
to complicate the game by constructing new rules on 


these initial data. Instead of which girls, though they 
are very fond of this game and play it much oftener than 
boys, have applied all their ingenuity in inventing new 
figures. For the game of Marelle exists in a multitude 
of forms ; the sections drawn in chalk on the pavement 
succeed one another in a straight line, in parallel lines, 
in the shape of a spiral, a circle, an oval, of the pipe of 
a stove, etc. But each game in itself is very simple and 
never presents the splendid codification and complicated 
jurisprudence of the game of marbles. As to the game 
of marbles itself, the few little girls who take any interest 
in it seem more concerned with achieving dexterity at 
the game than with the legal structure of this social 

As the extremely polymorphous nature of the game of 
Marelle made any interrogatory on rale consciousness 
difficult, we decided to study a very simple game con- 
taining a minimum of rules and to try and find out up 
to what point girls look upon rules as obligatory. In 
this case, as in those already dealt with, what interests 
us of course is to see what types of obligation appear at 
different ages and whether the youngest children are 
those who are most hostile to any alteration in the social 
heritage. As the game is simple and only girls are 
questioned, the conditions before us are as different as 
possible from the boys' game of marbles. Such analogies 
as do eventually appear will therefore be all the more 

The game of " ilet cachant " is one of the most primitive 
forms of the game of hide-and-seek. The little girl who is 
ilet (the derivation of the word seems to be " il est ", 
as the expressions " ilet courant ", " ilet cachant ", etc., 
seem to suggest) remains at a spot called the " tauche " 
( = the place one touches) while the others hide. Once 
the signal has been given the girl who is " ilet " begins 
to look for the others who try to reach the " tauche " 
before being caught. Whoever is caught is " ilet " for 
the next game. The first to play "ilet" is selected 


according to the well-known ritual formula, " Une boule 
deux boules trois boules roulent ! " etc. The little 
girls call this " plunging ". " The last to plunge is llet." 

The game being so simple, we shall not waste time by 
describing the stages during which it is learnt and the 
rules are put into practice. It will be sufficient to dis- 
tinguish two stages one before the age of 7 and one 
which extends from 6-7 years to 11-12. 

During the first of these stages, which here again we 
can designate as the stage of egocentrism, the children 
take great pleasure in imitating the ordered doings of 
their elders, but in practice know nothing of their raison 
d'etre ; each plays essentially for himself, just for the fun 
of running about or hiding, and above all, so as to do the 
same as the others. 

JACQUELINE (5 ; 7) is initiated into the game of " ilet " 
by an older friend (10 yrs.) who has immense prestige in 
her eyes, and she plays with this friend and with a few 
children of 8 to 12. As long as the game lasts (about 
three-quarters of an hour) she runs and hides with 
apparently the greatest enjoyment but without under- 
standing the point of the " tauche ". As soon as one of 
the children runs towards " home " crying " I've touched 
it" Jacqueline also runs to touch it ritually but without 
any relation to the other's conduct. She is quite happy 
playing her own little game on the fringe of the real game. 

The following days she behaves in the same way. 
Several days later she plays for half an hour, alone with 
the friend who has initiated her. This is what she does : 
i she still does not understand why you touch or hit 
the " tauche " but goes on touching it as soon as her 
playmate does (which is quite senseless, since the other 
child touches home to escape from her) ; 2 while she is 
waiting for her playmate to hide, she cheats in all good 
faith (she looks round on the sly, pretending to cover her 
face with her hands, she asks me for information who 
am simply an onlooker, etc.) ; 3 she enjoys losing quite 
as much as winning, her sole aim being to do the same as 
her older friend, though all the time she is running and 
hiding and shouting entirely on her own. 

The analogy strikes one at once between this and the 
behaviour of contemporary marble players : imitation of 


seniors mixed with egocentric play, no competition, no 
mutual control in the matter of rules. The child knows 
that there are rules and respects their external manifesta- 
tion : thus Jacqueline insists upon touching the " home " 
ritually, because she feels that this is an obligatory rite 
in the serious performance of the game of " ilet ". But 
this is far more participation in the life of the older 
children than an effort to cooperate with them. As for 
the application of rules, it allows for any amount of 
individual caprice (there is not even any consciousness of 
cheating) since the aim of the game is not yet social in 
the sense of ordered competition. 

But after the age of 6-7 in the average the child changes 
in his attitude and begins to observe rules. What matters 
to him now is not so much to do what the older ones 
are doing, though still acting entirely for himself, as beat- 
ing his partners by doing exactly the same thing as they. 
Hence the appearance of mutual control in the application 
of the law, together with an effective respect for obliga- 
tions (not to cheat when one is " he ", etc.). We shall call 
this the stage of cooperation. 

The next point to settle is, what are the feelings which 
the little girls entertain towards rules ? When we suggest 
some modification in the accepted usages, shall we meet 
with an opposition that increases as the child grows 
older, or shall we find that girls, like boys, gradually 
subordinate rules to mutual agreement and abandon the 
absolutely binding element in tradition ? 

The facts give unmistakable answer, though they point 
to a slight difference from what we observed in boys. 
Where the analogy is complete is that girls also begin 
by regarding the law as untouchable and innovations as 
illegitimate, and admit later on that rules become endowed 
with the force of law in so far as they are ratified by the 
collective will. On the other hand and this rather 
complicated our interrogatory, in relation to what we 
know about boys this change of orientation takes place 
on the average towards the age of 8, that is to say, it 


almost completely coincides with the inception of the 
cooperating stage. This early tolerance is clearly con- 
nected with the somewhat loosely knit character of the 
game of " tlet-cachant ". 

Roughly speaking, we can distinguish three stages. 
The first is contemporaneous with the beginnings of the 
game itself, with the first half, that is, of the egocentric 
stage. At this stage, the child seems ready to change 
all the rules and to show no inner respect for tradition 
and the example of its seniors. But, as we saw in the 
case of the game of marbles, this is so only in appearance, 
and the child accepts proposed modifications in so far 
as it believes them to correspond with earlier decrees 
Thus Jacqueline touches the " home " in so far as she 
sees this rite practised by others, but if anyone neglects 
to carry out this duty, she is in no way scandalized, 
deeming that this too is among the things that " are 
done ". There is no need to go over this period again, 
which in any case is a short one and very difficult to 
analyse for lack of any consistency in the children's 

During a second stage, extending on the average up to 
the end of the seventh year, the little girls we questioned 
showed themselves to be firmly attached to the prevailing 
usage. Also, like boys, they think that rules are of adult 
and quasi-divine origin. 

MOL (6J) : " Are there things you must do and things 
you mustn't do in this game ? Yes. The things you must 
do are the rules of the game. Could you invent a new 
rule ? . . . Supposing you said that the third who 
plunged was ' he ' ? Yes. Would it be all right to play 
that way or not ? It'd be all right. Is it a fair rule like 
the others ? Less fair. Why ? Because the last one has 
to be ' he '. And if everyone played that way, would it 
become a fair rule ? No. Why ? Because the game isn't 
like that. How did the rules begin ? . . . How do 
you know how to play ? / learn 'd the first time. I didn't 
know how. We played with a little girl who told us. 
And how did the little girl know ? She learn' d. And 
did your Mummy play when she was little ? Yes. The 


school-teacher taught her. -But how did it begin ? People 
learn' d with the school-teacher, Who made up the rules, 
grown-ups or children ? Grown-ups. And if a child 
invents a new rule, is that all right ? No it isn't. Why 
not ? Because they don't know how to." 

AGE (7 yrs.) admits that the third child to be caught 
could be regarded as '' he ", but she refuses to recognize 
this new rule as fair. " Is it a real rule ? Yes. Is it 
fair ? jYo. Became that is not the way you play." (The 
conversation is resumed after break.) " Would that new 
rule be all right ? No, because the first one who is caught 
has to be { he '. Why ? Because otherwise it wouldn't be 
fair. But if everyone played like that ? It wouldn't 
do because the third mustn't le* he'! 9 

BON (7 yrs.) admits that her companions would be 
pleased with such an innovation, but this would be 
" not fair. Why ? Because it upsets the whole game, 
"because it's wrong/' 

Ros (8J) invents a new rule : " You might say that 
only one goes and hides and then the others would go and 
look for her. Would that be all right for playing ? Yes. 
Is this new way fairer than the other way or less fair ? 
Less fair. Why ? Because you don't play with that 
onCf But if everyone played that way ? Then it would 
be just a little bit fair. Why a little bit ? Because^ after 
all, it is a little the same way. But it isn't quite fair ? 
jVo. Why? . . . And if everyone agreed to play 
that way would it be the same, or would it not ? It 
would be all right (reluctantly). Which is best, always to 
play the same way or to change ? To change (still 
reluctantly). Why ? Because that game is truer. Which 
one ? Not the one I made up. Then is it best to change 
it or to leave it as it is IBest to leave it." The game 
of " llet cachant " has been invented by " a Gentleman ". 
" Has it been changed since then, or is it the same as 
when it was invented? Yes, it hasn't been changed. 
But if people want to, they can change it ? Yes. Can 
children? Yes. If they invent something will it be 
more fair or less fair ? Less fair. Why ? Because it 
isn't the real game. What is the real game ? The one 
the gentleman invented. Why ? Because that is the one 
you always play." 

LIL (8 ; 10) : " How did the game begin the very first 
time of all ? I think a lady invented it. Do yoii think 
the ga$ie has changed since then ? People may have 


changed it. Who changed it, grown-ups or children ? 
Grown-ups. But do you tMnk children can change it ? 
Yes, they can. Could you & for instance, if you wanted 
to ? Oh, if I wanted to, yes. Would it do or would it 
not do ? It would do just as well. Would your friends 
be pleased ? They'd be just as pleased. Would it be as 
fair or less fair ? Less fair. Fair in what way ? 
I think the lady who invented it invented it better. Why ? 
Because grown-ups are cleverer, because they have been at 
school longer than children." 

These examples show that girls of this stage, while 
perhaps not quite so keen on conformity as boys, show 
a sufficient feeling for tradition to ensure respect for 
rules. We even come across the divine right of mothers 
to lay down the laws of " ttet cachant ". Buc tells us at 
6 years old that " God taught them ?; . 

But at the age of 8 a good half of the girls we questioned 
have changed their attitude and declare that the new 
rule is as good as the old, provided it is practicable and, 
above all, provided it rallies all the votes. It is on this 
point that the girls, more tolerant and more easily recon- 
ciled to innovations, struck us as being slightly different 
from the boys. 

BAG (10 ; 4) is asked to judge a new rule suggested 
by one of her companions and which consists in not 
" struggling " when you are caught. " Is it just as fair 
or less fair ? Just as fair. Is it a real rule or not ? 
A real rule. What is a real rule? It's something you 
play at really and truly. But no one has played yet with 
the rule invented by your friend ; is it real all the same, 
or not ? It is real. Would it work ? Yes. Would your 
friends be pleased or not ? Not pleased, because they'd 
never want to be 'he' (in that way). And if they agreed 
about it, would it be fair or not ? Yes." 

CHO (9 yrs.) : " Is this rule as real as the others ? 
No. Why ? Because you never play that way. Would 
it be all right to play that way? Yes. Would your 
friends like it ? Yes. Is this new rule more fair or less 
fair than the other ? Both the same. Which is the most 
real? Both the same. How did rules begin? Some- 
body invented them. Who ? A child . . . There were 


some kids who were playing and the others did the same 
iking. Were the rales those children made fair ? Yes. 
When is a rule fair ? W hen it's all right for playing. 
When is it real ? When it's all right for playing" 

These reactions, which are characteristic of what we 
found in girls, are thus both like and unlike those of boys. 
They are alike in so far as cooperation between the 
players brings about the gradual diminution of rule 
mysticism : the rule is no longer an imperative coming 
from an adult and accepted without discussion, it is a 
means of agreement resulting from cooperation itself. 
But girls are less explicit about this agreement and this is 
our reason for suspecting them of being less concerned 
with legal elaborations. A rule is good so long as the 
game repays it. 

Little girls are therefore extremely tolerant, and it 
never occurs to them to introduce a distinguo and to 
codify the possible cases or even the very conditions of 

Is this difference due to the somewhat loosely-knit 
"character of the game of " ilet cachant " or to the actual 
mentality of little girls ? Both these suppositions probably 
amount to the same thing, since we noticed that all 
girls' games are marked by this polymorphism and 
tolerance. The question, moreover, does not interest us 
here, and it is not this contrast which we propose to study. 
All that needs to be emphasized from the point of view 
of the psychology of rules is that, in spite of these 
differences in the structure of the game -an^ apparently, 
in the players' mentality, we find the sine process at 
work as in the evolution of the game of marbles : first 
a mystical respect for the law, which is conceived as 
untouchable and of transcendental origin, then a co- 
operation that liberates the individuals from their practical 
egocentrism and introduces a new and more immanent 
conception of rules. 

KINDS OF RESPECT. In order to pursue our analysis 


with any profit we shall have to draw from the material 
that has been presented above certain conclusions which 
will serve as guiding hypotheses in the chapters that are 
to follow. In other words, we shall try to find in the 
various stages we have examined certain evolutionary 
processes which are likely to reappear in our future 

Two prefatory questions confront us. The first has 
to do with differences of structure and differences of 
degree. Rules evolve as the child grows older ; neither 
the practice nor the consciousness of rules is the same 
at six as it is at twelve. Is the difference one of nature 
or of degree ? After having done our utmost to show 
that child thought differs from adult thought not only in 
degree but in its very nature, we confess that we no longer 
know precisely what is meant by these terms. From the 
methodological point of view their meaning is, of course, 
perfectly clear ; they tell us to beware of facile analogies 
and to look for the less obvious differences before pointing 
to resemblances that will stand out of themselves. But 
how does the matter stand from the theoretical point of 
view ? Psychologically, as M. Bergson has well shown, 
every difference of degree or quantity is also a difference 
of quality. Conversely, it is difficult to think of a 
difference in kind without the presence of at least some 
functional continuity, so that between two successive 
structures a succession of intermediate degrees could be 
found. For instance, after having tried to describe the 
child's mentality as distinct from the adult's we have 
found ourselves obliged to include it in our descriptions 
of the adult mind in so far as the adult still remains a 
child. This happens particularly in the case of moral 
psychology, since certain features of child morality always 
appear to be closely connected with a situation that 
from the first predominates in childhood (egocentrism, 
resulting from the inequality between the child and the 
adult surrounding which presses upon him) but which 
may recur in adult life, especially in the strictly conformist 


and gerontocratic societies designated as primitive. Con- 
versely, in certain circumstances where he experiments in 
new types of conduct by cooperating with Ms equals, 
the child is already an adult. There is an adult in every 
child and a child in every adult. The difference in nature 
reduces itself to this. There exist in the child certain 
attitudes and beliefs which intellectual development will 
more and more tend to eMminate : there are others 
which will acquire more and more importance. The 
latter are not simply derived from the former but are 
partly antagonistic to them. The two sets of phenomena 
are to be met both in the child and in the adult, but 
one set predominates in the one, the other in the 
other. It is, we may say, simply a question of the 
proportions in which they are mixed ; so long as we 
remember that every difference of proportion is also a 
difference of general quality, for the spirit is one and 

Between the various types of rules which we shall give 
there will therefore be at once continuity and qualitative 
difference : continuity of function and difference of 
structure. This renders arbitrary any attempt to cut 
mental reality up into stages. The matter is further 
complicated by the " Law of conscious realization " 
and the resulting time-lag. The appearance of a new 
type of rule on the practical plane does not necessarily 
mean that this rule will come into the subject's con- 
sciousness, for each mental operation has to be re- 
learnt on the different planes of action and of thought. 
There are therefore no inclusive stages which define the 
whole of a subject's mental life at a given point of his 
evolution ; the stages should be thought of as the 
successive phases of regular processes recurring like a 
rhythm on the superposed planes of behaviour and of 
consciousness. A given individual may, for example, 
have reached the stage of autonomy with regard to a 
certain group of rules, while his consciousness of these 
rules, together with the practice of certain more subtle 


rules, will still be coloured with heteronomy. 1 We cannot 
therefore speak of global or inclusive stages characterized 
as such by autonomy or heteronomy, but only of phases of 
heteronomy or autonomy which define a process that is 
repeated for each new set of rules or for each new plane of 
thought or reflection. 

A second prefatory question faces us : that of society 
and the individual. We have sought to contrast the child 
and the civilized adult on the ground of their respective 
social attitudes. The baby (at the stage of motor intelli- 
gence) is asocial, the egocentric child is subject to external 
constraint but has little capacity for cooperation, the 
civilized adult of to-day presents the essential character of 
cooperation between differentiated personalities who regard 
each other as equals. There are therefore three types of 
behaviour : motor behaviour, egocentric behaviour (with 
external constraint), and cooperation. And to these three 
types of social behaviour there correspond three types of 
rules : motor rules, rules due to unilateral respect, and rules 
due to mutual respect. But here again, one must beware 
of laying down the law : for things are motor, individual 
and social all at once. As we shall have occasion to show, 
rules of cooperation are in some respects the outcome of 
the rules of coercion and of the motor rules. On the 
other hand, coercion is applied during the first days of 
an infant's life, and the earliest social relations contain 
the germs of cooperation. Here again, it is not so much 
a question of these successive features themselves as of 
the proportions in which they are present. Moreover, 
the way in which conscious realization and the time-lag 
from one level to another come into play is a further bar 
to our arranging these phenomena in a strict sequence, 
as though they made a single appearance and then dis- 
appeared from the scene once and for all. 

1 A child of 10 will, for example, show signs of autonomy in his 
application of the rules of the game of marbles, but will give proof of 
heteronomy in the extent to which he is conscious of these rules and 
in his application of rules relating to lying and justice. 


With these reservations in mind, let us then try to 
outline the processes which govern the evolution of the 
idea of rules. And if language and discursive thought 
which, according to a famous metaphor, are necessarily 
cinematographic in character, tend to lay too much 
emphasis on discontinuity, let it be understood once and 
for all, that any over-sharp discontinuities are analytical 
devices and not objective results. 

To continue, our enquiry into the nature of games 
would seem to reveal the existence of three types of rules, 
and the problem before us will be to determine the exact 
relations between them. There is the motor rule, due to 
preverbal motor intelligence and relatively independent 
of any social contact ; the coercive rule due to unilateral 
respect ; and the rational rule due to mutual respect. 
Let us examine these three rules in succession. 

The motor rule. In its beginnings the motor rule 
merges into habit. During the first few months of an 
infant's life, its manner of taking the breast, of laying 
its head on the pillow, etc., becomes crystallized into 
imperative habits. This is why education must begin in 
the cradle. To accustom the infant to get out of its own 
difficulties or to calm it by rocking it may be to lay the 
foundations of a good or of a bad disposition. But not 
every habit will give rise to the knowledge of a rule. 
The habit must first be frustrated, and the ensuing con- 
flict must lead to an active search for the habitual. 
Above all, the particular succession must be perceived 
as regular, i.e. there must be judgment or consciousness 
of regularity (Regelbewusstseiri). The motor rule is there- 
fore the result of a feeling of repetition which arises 
out of the ritualization of schemas of motor adaptation. 
The primitive rules of the game of marbles (throwing 
the marbles, heaping them, burying them, etc.) which 
we observed towards the age of 2-3 are nothing else. 
The behaviour in question starts from a desire for a form 
of exercise which takes account of the particular object 
that is being handled. The child begins by incorporating 


the marbles Into one or other of the schemas of assimilation 
already known to him, such as making a nest,, hiding 
under earth, etc. Then he adapts these schemas to the 
nature of the object by preventing the marbles from 
rolling away by putting them In a hole, by throwing them, 
etc. This mixture of assimilation to earlier schemas and 
adaptation to the actual conditions of the situation is 
what defines motor intelligence. But and this is where 
rules come into existence as soon as a balance Is estab- 
lished between adaptation and assimilation, the course 
of conduct adopted becomes crystallized and ritualized. 
New schemas are even established which th,e child looks 
for and retains with care, as though they were obligatory 
or charged with efficacy. 

But is this early behaviour accompanied by conscious- 
ness of obligation or by a feeling of the necessity of the 
rule ? We do not think so. Without the feeling of 
regularity which goes to the formation of any intelligence 
and already so clearly characterizes motor intelligence, 
the consciousness of obligation would no doubt never 
make its appearance. But there is more in this conscious- 
ness of obligation than a mere perception of regularity, 
there is a feeling of respect and authority which, as 
Durkheim and Bovet have clearly shown, could not come 
from the individual alone. One might even be tempted 
to say that rules only begin when this consciousness of 
obligation, i.e. when the social element has made its 
appearance. But the material we have collected all goes 
to show that this obligatory and sacred character only 
marks an episode in the evolution of rules. After being 
unilateral, respect becomes mutual. In this way, the rule 
becomes rational, i.e. it appears as the fruit of a mutual 
engagement. And what is this rational rule but the 
primitive motor rule freed from individual caprice and 
submitted to the control of reciprocity ? 

Let us therefore turn to the influence of inter-individual 
relations in the constitution of rules. In the first place, 
we repeat, the social element is to be found everywhere. 


From the hour of its birth certain regularities of conduct 
are imposed upon the infant by the adult, and, as we 
have shown elsewhere (C.W. and C.C.), every regularity 
observed in nature, every " law " appears to the child 
for a long time as both physical and moral. Even in 
connection with the preverbal stage, characterized as it 
is by the motor rule in all its purity, people have spoken 
of child "sociology''. Thus Mme Ch. Biihler, in her 
interesting studies on the first year, has very accurately 
noted how much more interested a baby is in people than 
in things. Two considerations, however, forbid us to 
regard these facts as playing a very important part in 
the development of motor rules. In the first place, a 
baby, as Mme Biihler has acutely noted, is more interested 
in adults than in its contemporaries. Now, surely this 
shows either that an interest in what is big, powerful, 
and mysterious (to say nothing of the interest in food and 
physical comfort bound up with the person of the parents) 
still outweighs any social interest, or what perhaps 
comes to the same thing that inter-individual relations 
based on admiration and unilateral respect are stronger 
than relations based on cooperation. In either case, a 
baby of 10-12 months, which elaborates all sorts of 
ritual acts connected with the objects it handles, may 
be influenced indirectly by its feelings for the adult, but 
neither the baby nor anyone observing it could distinguish 
these influences from the rest of what constitutes its 
universe. But the same child at about two, once he is 
able to speak or to understand what is said to him, will 
be acutely conscious of the rules that are imposed upon 
him (sitting down to meals or going to bed when he 
would like to go on playing) and will distinguish them 
perfectly well from the motor rules or rituals which he has 
himself established in the course of his games. It is the 
increasing constraint exercised upon the child by those 
around him that we regard as the intervention of the 
social factor. 

In the case of play rules, the discontinuity between 


this process and the purely motor processes Is obvious. 
On a given occasion, the child meets with others, older 
than himself, who play marbles according to a code. 
Immediately he feels that he ought to play in the same way 
himself ; immediately he assimilates the rules adopted in 
this way to the totality of commands which control his 
way of living. In other words, he immediately places the 
example of children older than himself on the same 
plane as the hundred and one other customs and obliga- 
tions imposed by adults. This is not an explicit process 
of reasoning. The child of three or four is saturated with 
adult rules. His universe is dominated by the idea that 
things are as they ought to be, that everyone's actions 
conform to laws that are both physical and moral, in a 
word, that there is a Universal Order. The revelation 
of the rales of the game, of " the real game " as played 
by his seniors is immediately incorporated into this 
universe. A rule imitated in this way is felt from the 
first as something obligatory and sacred. 

Only, the main result of our enquiry, and one which 
will receive repeated confirmation in the latter part of 
this book, is that the social factor is not just one thing. 
If there is relative discontinuity between the early motor 
activity and adult intervention, the discontinuity is no 
less marked between the unilateral respect which accom- 
panies this intervention and the mutual respect which 
gradually comes into being later on. Once again, let 
there be no misunderstanding : the qualities in question 
are not more important than the proportions in which 
they are mixed. Between the unilateral respect of the 
little child who receives a command without even the 
possibility of disagreement and the mutual respect of 
two adolescents who exchange their points of view there 
is room for any number of intermediate stages. Constraint 
is never unadulterated, nor, therefore, is respect ever 
purely unilateral : the most submissive child has the 
feeling that he can, or could, argue that a mutual sympathy 
surrounds relationships that are most heavily charged 


with authority. And conversely, cooperation is never 
absolutely pure : in any discussion between equals, one 
of the disputants can always exert pressure on the other 
by making overt or hidden appeals to custom and 
authority. Cooperation, indeed, seems rather to be the 
limiting term, the ideal equilibrium to which all relations 
of constraint tend. As the child grows up, his relations 
with adults approximate to equality, and as communities 
develop, their group ideas leave more room for free 
discussion between individuals. Nevertheless, every time 
the proportion of constraint and cooperation is changed, 
mental states and conduct are marked by a correspond- 
ingly fresh quality, so that, however artificial the analysis 
may seem, it is necessary to distinguish these two processes 
as leading to different results. 

Let us begin with unilateral respect and the coercive rule 
to which it leads. The outstanding fact here, and what 
differentiates this type of respect from its successor, is 
the close connection which we have noted between respect 
due to the constraint of older children or adults and the 
egocentric behaviour of the child between 3 and 7. Let 
us therefore examine this point afresh in order to establish 
its general significance. 

The facts are, it will be remembered, as follows. On 
the one hand, the child knows that there are rules, the 
"real rules", and that they must be obeyed because 
they are obligatory and sacred ; but on the other hand, 
although the child vaguely takes note of the general 
scheme of these rules (making a square, aiming at the 
square, etc.) he still plays more or less as he did during 
the previous stage, i.e. he plays more or less for himself, 
regardless of his partners, and takes more pleasure 
in his own movements than in the observance of the 
rules themselves, thus confusing his own wishes with 

The right interpretation of these facts calls for very 
close scrutiny, so easy is it to fall into mistakes in 
dealing with the problem of the socialization of the child. 


In the first place, let us remind the reader that the behaviour 
of children of 3 to 7 with regard to the game of marbles 
is comparable on aH points to the behaviour of children 
of the same age in regard to their conversations or to 
their social and intellectual life in general But the 
egocentrism common to all these types of behaviour 
admits of at least two interpretations. Some people 
think and in all our previous works we have claimed 
to be of their number that egocentrism is presocial in 
the sense that it marks a transition between the individual 
and the social, between the motor and quasi-solipsistic 
stage of the baby and the stage of cooperation proper. 
However closely connected with unilateral respect ego- 
centrism may be, this mixture of coercion and subjectivity 
which characterizes the stage lasting from 2 to 7 years 
does seem to us less social than cooperation (which is 
the one determining factor in the formation of the rational 
elements in ethics and in logic). Other thinkers, on the 
contrary, consider egocentric behaviour to be in no way 
presocial the social element remaining identical with 
itself throughout all the various stages but take it to be, 
as it were, parasocial behaviour, analogous to what occurs 
in the adult when private feeling obscures his objectivity 
or when he is left out of a conversation from which he is 
precluded by his incompetence or stupidity. 1 Thinkers 
belonging to this second group can see no essential 
difference between cooperation and coercion ; hence their 
view that the social factor is a permanent element through- 
out the whole course of mental development. 

The data with which the present discussion is concerned 
would seem to be of a nature to remove these ambiguities. 
Egocentrism is both presocial, in view of the eventual 
cooperation, and parasocial, or simply social, in relation 
to the constraint of which it constitutes the most direct 

1 See Blondel, " Le Langage et la Pensde chez I'enfant d'aprfcs 
un livre recent." Revue Hisi. Phil. Rel. (Strasburg), Vol. IV (1924), 
p. 474 et seq. 


To understand this we need only analyse the relations 
of the younger to the older children. Every observer 
has noted that the younger the child, the less sense he 
has of Ms own ego. From the intellectual point of view, 
he does not distinguish between external and internal, 
subjective and objective. From the point of view of 
action, he yields to every suggestion, and if he does 
oppose to other people's wills a certain negativism which 
has been called " the spirit of contradiction " l this only 
points to his real defencelessness against his surroundings. 
A strong personality can maintain itself without the 
help of this particular weapon. The adult and the 
older child have complete power over him. They impose 
their opinions and their wishes, and the child accepts 
them without knowing that he does so. Only and this is 
the other side of the picture as the child does not dis- 
sociate his ego from the environment, whether physical 
or social, he mixes into all his thoughts and all his actions, 
ideas and practices that are due to the intervention of 
his ego and which, just because he fails to recognize them 
as subjective, exercise a check upon his complete socializa- 
tion. From the intellectual point of view, he mingles his 
own fantasies with accepted opinions, whence arise pseudo 
lies (or sincere lies), syncretism, and all the features of 
child thought. From the point of view of action, he 
interprets in his own fashion the examples he has adopted, 
whence the egocentric form of play we were examining 
above. The only way of avoiding these individual re- 
fractions would lie in true cooperation, such that both 
child and senior would each make allowance for his own 
individuality and for the realities that were held in common. 

But this presupposes minds that know themselves and 
can take up their positions in relation to each other. It 
therefore presupposes intellectual equality and reciprocity, 
both of them factors that are not brought about by 
unilateral respect as such. 

1 See Mme Reynier, " L'Esprit de contradiction chez Venfant," 
La Nouvelle Education, V. 1926, pp. 45^52- 


Egocentrism in so far as it means confusion of the ego 
and the external world, and egocentrism in so far as it 
means lack of cooperation, constitute one and the same 
phenomenon. So long as the child does not dissociate 
Ms ego from the suggestions coming from the physical 
and from the social world, he cannot cooperate, for 
in order to cooperate one must be conscious of one's 
ego and situate it in relation to thought in general And 
in order to become conscious of one's ego, it is necessary 
to liberate oneself from the thought and will of others. 
The coercion exercised by the adult or the older child 
is therefore inseparable from the unconscious egocentrism 
of the very young child. 

If, now, we turn to children's societies below the age 
of 8, we shall constantly meet with phenomena of this 
order. No setting seems so favourable to the contagion 
and even the constraint of the older ones than these 
early societies ; not a gesture of the little ones but has 
been, as it were, commanded or suggested to them. 
We have not here any autonomous individuals, any con- 
scious minds that impose themselves in virtue of an 
inner law to which they themselves are subject. And 
yet there is far less unity, far less real cooperation than 
in a society of 12-year-olds. Egocentrism and imitation 
are one, 1 and the same applies later on to autonomy and 
cooperation. It is therefore no mere chance that nearly 
all little children assimilate the rules learned in these 
surroundings to the moral rules imposed by adults and 
by the parents themselves. 

It is perhaps possible to go further and to connect 
egocentrism with the belief in the divine origin of institu- 
tions. Childish egocentrism is in its essence an inability 
to differentiate between the ego and the social environ- 
ment. Now the result of this non-differentiation is that 
the mind is unwittingly dominated by its own tendencies, 
in so far as these are not diminished or rendered 

i See L.T., p. 41. 


conscious by cooperation. But at the same time, all 
the opinions and commands that are adopted appear to 
be endowed with a transcendental origin. We have 
already ( 5) drawn attention to the very significant 
difficulty experienced by very young children in dis- 
tinguishing between what they have invented themselves 
and what has been imposed upon them from outside. 
The mind's content is felt both as very familiar and as 
superpersonal, permanent and in a sense revealed. Nothing 
is more characteristic of childhood memories than this com- 
plex sensation of gaining access to one's most intimate 
possessions and at the same time of being dominated by 
something greater than oneself which seems like a source of 
inspiration. There is little mysticism without an element 
of transcendence, and, conversely, there is no transcendence 
without a certain degree of egocentrism. It may be that the 
genesis of these experiences is to be sought in the unique 
situation of the very young child in relation to adults. The 
theory of the filial origin of the religious sense seems to us 
singularly convincing in this connection. 

To return, however, to our analysis of the game of 
marbles, it is a highly significant fact that it is the 
younger and not the older children who believe in the 
adult origin of rules, although they are incapable of 
really putting them into practice. The belief here is 
analogous to that prevalent in conformist communities, 
whose laws and customs are always attributed to some 
transcendental will. And the explanation is always the 
same. So long as a practice is not submitted to conscious, 
autonomous elaboration and remains, as it were, external 
to the individual, this externality is symbolized as trans- 
cendence. Now in the case of the child, exteriority and 
egocentrism go hand in hand in so far as egocentrism is 
preserved by the constraint exercised from outside. If, 
therefore, the children of the earlier stages were those 
who showed the maximum respect for rules together 
with the most pronounced belief in their transcendental 
origin, this was not due to any fortuitous resemblance. 


The two features coexisted In virtue of an inner logic 
which is the logic of unilateral respect. 

Let us now deal with mutual respect and rational rules. 
There is, in our opinion, the same relation between mutual 
respect and autonomy as between unilateral respect 
and egocentrism, provided the essential qualification be 
added, that mutual respect far more than unilateral 
respect, joins forces with the rationality already incipient 
in the motor stage, and therefore extends beyond the 
phase that is marked by the intervention of constraint 
and egocentrism. 

We have, in connection with the actual facts examined, 
pointed to the obvious correlation between cooperation 
and the consciousness of autonomy. From the moment 
that children really begin to submit to rules and to apply 
them in a spirit of genuine cooperation, they acquire a 
new conception of these rules. Rules become something 
that can be changed if it is agreed that they should be, 
for the truth of a rule does not rest on tradition but on 
mutual agreement and reciprocity. How are these facts 
to be interpreted ? In order to understand them, all we 
have to do is to take as our starting-point the functional 
equation uniting constraint and egocentrism and to take 
the first term of the equation through the successive 
values which link up constraint and cooperation. At the 
outset of this genetic progression, the child has no idea of 
his own ego ; external constraint works upon him and he 
distorts its influence in terms of his subjectivity, but he 
does not distinguish the part played by his subjectivity 
from that played by the environmental pressure. Rules 
therefore seem to him external and of transcendental 
origin, although he actually fails to put them into practice. 
Now, in so far as constraint is replaced by cooperation, 
the child dissociates his ego from the thought of other 
people. For as the child grows up, the prestige of older 
children diminishes, he can discuss matters more and 
more as an equal and has increasing opportunities (beyond 
the scope of suggestion, obedience, or negativism) of freely 


contrasting Ms point of view with that of others. Hence- 
forward, he will not only discover the boundaries that 
separate Ms self from the other person, but will learn to 
understand the other person and be understood by him. 
So that cooperation is really a factor in the creation of 
personality, if by personality we mean, not the unconscious 
self of childish egocentrism, nor the anarcMcal self of 
egoism in general, but the self that takes up its stand on the 
norms of reciprocity and objective discussion, and knows 
how to submit to these in order to make itself respected. 
Personality is thus the opposite of the ego 1 and this 
explains why the mutual respect felt by two personalities 
for each other is genuine respect and not to be confused 
with the mutual consent of two individual " selves " 
capable of joining forces for evil as well as for good. 
Cooperation being the source of personality, rales cease, in 
accordance with the same principle, to be external They 
become both the constitutive factors of personality and its 
fruit, in accordance with the circular process so frequently 
*empMed in the course of mental development. In tMs way 
autonomy succeeds heteronomy. 

TMs analysis will have shown how new in quality are 
the results of mutual respect as compared with those 
that arose out of unilateral respect. And yet the former 
is the outcome of the latter. Mutual respect is, in a 
sense, the state of equilibrium towards which unilateral 
respect is tending when differences between child and 
adult, younger and older are becoming effaced ; just as 
cooperation is the form of equilibrium to wMch con- 
straint is tending in the same circumstances. In spite 
of this continuity in the facts it is necessary, nevertheless, 
to distinguish between the two kinds of respect, for their 
products differ as greatly as do autonomy and ego- 

It can even be maintained that mutual respect and 
cooperation are never completely realized. They are not 

i See Ramon Fernandez, De la Personnalitt, Au Sans Pareil (Paris), 


only limiting terms, but ideals of equilibrium. Every- 
where and always the quota of generally accepted rules 
and opinions weighs, however lightly, on the individual 
spirit, and it is only in theory that the child of 12-14 
can submit all rales to a critical examination. Even the 
most rational of adults does not subject to Ms " moral 
experience ?J more than an infinitesimal proportion of the 
rules that hedge Mm round. Anxious though he was to 
escape from Ms " provisional morality ", Descartes retained 
it to the end of Ms days. 

But we are not concerned with the question as to 
whether cooperation is ever completely realized or whether 
it remains only a theoretical ideal. Psychologically, the 
same rale is a completely different reality for the child 
of 7 who regards it as sacred and untouchable and for the 
child of 12 who, without interfering with it, regards it 
as valid only after it has been mutually agreed upon. 
The great difference between constraint and cooperation 
or between unilateral respect and mutual respect, is that 
the first imposes beliefs or rules that are ready made and 
to be accepted en bloc, while the second only suggests a 
method a method of verification and reciprocal control 
in the intellectual field, of justification and discussion in 
the domain of morals. It matters little whether this 
method be applied immediately to all the rules imposed 
by the environment or only to one aspect of behaviour : 
once it has come into existence it has the right to be 
applied to everything. 

TMs fundamental difference between constraint and 
cooperation (the one laying down ready-made rules, the 
other giving a method for the elaboration of rules) will 
supply us straight away with an answer to an objection 
wMch is bound to crop up in the course of our analysis of 
the products of mutual respect. Supposing, it will be said, 
that mutual respect does constitute the essential factor 
in the behaviour of children of 12-13 and over, how can 
we attribute to it a genuinely moral effect ? It is easy 
enough to see that mutual consent is sufficient to explain 


the establishment of rules of the game, since the child is 
urged to play both by interest and by pleasure. But 
when we come to actual moral rules (not to lie, not to 
steal, etc.), why is it that mutual respect does not make 
the children come to some agreement on the subject of 
what adults consider to be wrong ? Take a band of 
young ruffians whose collective activity consists in thieving 
and in playing practical jokes on honest folk ; is not the 
mutual consent subsisting between its members com- 
parable, psychologically, to the mutual respect that holds 
between marble players ? Now, apart from the fact that 
there is honour among thieves, this difficulty can easily 
be disposed of. In the first place, a distinction should, 
as we have seen, be drawn between mutual consent in 
general and mutual respect. There may be mutual consent 
in vice, since nothing will prevent the anarchical tendencies 
of one individual from converging with those of another 
individual. Whereas the word " respect " implies (at 
least as regards mutual respect) admiration for a per- 
sonality precisely in so far as this personality subjects 
itself to rules. Mutual respect would therefore seem to 
be possible only within what the individuals themselves 
regard as morality. 

Moreover, as soon as cooperation comes into being (in 
the moral as well as in the intellectual field) "one must 
distinguish between the method and its results, or, as a 
contemporary logician has so cogently put it, between 
" constitutive reason " (practical or theoretical) and 
" constituted reason ". There are two kinds of rules, 
""those that are constitutive and render possible the exercise 
of cooperation, and those that are constituted and are 
the result of this very exercise. We have already been 
led to make this distinction in connection with the rules 
of a game. The rules of the Square, of the Coche, etc., 
which are observed by children of 11-13 are " constituted " 
rules, due to mutual consent and capable of being altered 
by general opinion. The precedence given to justice as 
opposed to chance, on the other hand, of effort over easy 


gain are " constitutive " rales, for without this " spirit 
of the game " no cooperation would be possible. In the 
same way, so-called moral rules can, generally speaking, 
be divided into constituted rules dependent upon mutual 
consent, and constitutive rules or functional principles 
which render cooperation and reciprocity possible. But 
how can these constitutive rules be regarded as them- 
selves the outcome of mutual respect since they are 
necessary to the latter's formation ? The difficulty here 
is purely formal. Between mutual respect and the rules 
which condition it there exists a circular relation analogous 
to that which holds between organ and function. Since 
cooperation is a method, it is hard to see how it could 
come into being except by its own exercise. No amount 
of constraint could determine its emergence. If mutual 
respect does derive from unilateral respect, it does so by 

We are faced, then, with three types of rules : the 
motor rule, the coercive rule founded on unilateral respect, 
and the rational rule (constituted or constitutive) due to 
mutual respect. We have outlined above the relation in 
which the last two types stand to each other. We have 
examined elsewhere how the first two succeed one another. 
It remains for us to show what are the relations of the 
rational rule to the motor rule. 

Generally speaking, one can say that motor intelligence 
contains the germs of completed reason. But it gives 
promise of more than reason pure and simple. From the 
moral as from the intellectual point of view, the child is 
born neither good nor bad, but master of his destiny. Now, 
if there is intelligence in the schemas of motor adaptation, 
there is also the element of play. The intentionality 
peculiar to motor activity is not a search for truth but 
the pursuit of a result, whether objective or subjective ; 
and to succeed is not to discover a truth. 

The motor rule is therefore a sort of experimental 
legality or rational regularity, and at the same time a 
play ritual. It will take one or other of these two forms 


according to circumstances. Now, at the moment when 
language and imagination are added to movement, ego- 
centrism directs the child's activity towards subjective 
satisfaction, while, at the same time, adult pressure 
imposes on Ms mind a system of realities which at first 
remains opaque and external. Constraint and egocentrism 
therefore interpose between motor intelligence and reason 
a complexus of realities which seem to interrupt the 
continuity of evolution. It is at this point that the 
motor rule is followed by the coercive rule, a crystallized 
social product which shows the sharpest contrast with the 
fragile tentative products of the initial motor intelligence, 
though, as we have seen, egocentric play continues in a 
sense the early gropings of the motor stage. 

But as the element of constraint is gradually eliminated 
by cooperation, and the ego is dominated by the per- 
sonality, the rational rule so constituted recaptures the 
advantages of the motor rule. The play of n-year-old 
children is in some ways closer to the motor accommodation 
of the one-year-old child in all its richness and truly 
experimental qualities than to the play of 7-year-olds. 
The boy of eleven plans Ms strokes like a geometrician 
and an artist in movement, just as the baby acts as a 
mechanician in handling objects and as an experimenter 
in inventing its rules. At the age of 6 or 7, on the contrary, 
the child is apt to neglect this element of invention, and 
to confine himself to imitation and the preservation of 
rites. But the immense superiority of the eleven-year-old 
player over the one-year-old, a superiority perhaps 
acquired by passing through the intermediate stage, is 
that Ms motor creations are no longer at the mercy of 
individual fantasy. The eleven-year-old has re-discovered 
the schema of experimental legality and rational regularity 
practised by the baby. But the motor rule found by the 
baby tends constantly to degenerate into play ritual, 
whereas the eleven-year-old invents notMng without the 
collaboration of his equals. He is free to create, but on 
condition of submitting to the norms of reciprocity. The 


motor being and the social being are one. Harmony is 
achieved by the union of reason and nature, whereas 
moral constraint and unilateral respect oppose super- 
nature to nature and mysticism to rational experiment. 

The discussion of the game of marbles seems to have 
led us into rather deep waters. But in the eyes of children 
the history of the game of marbles has quite as much 
"importance as the history of religion or of forms of 
government. It Is a history, moreover, that is magni- 
ficently spontaneous ; and it was therefore perhaps not 
entirely useless to seek to throw light on the child's 
judgment of moral value by a preliminary study of the 
social behaviour of children amongst themselves. 


THESIS. Before pursuing our analysis any further, it 
will be well to consider the results we have so far obtained 
in the light of the two principal hypotheses that have 
been brought forward concerning the psychological nature 
of respect and moral laws. If we refuse to accept Kant's 
view of respect as inexplicable from the point of view of 
experience, 1 only two solutions remain. Either respect is 
directed to the group and results from the pressure exercised 
by the group upon the individual or else it is directed to 
individuals and is the outcome of the relations of individuals 
amongst themselves. The first of these theses is upheld by 
Durkheim, the second by M. Bovet. The moment has not 
yet come for us to discuss these doctrines for their own sake, 
but at the same time we must, without anticipating our 
final critical examination, develop a working hypothesis 
that will take account of all possible points of view. This 
is all the more indispensable since the discrepancy between 
results obtained by these authors is chiefly due, as will 
be shown later on, to differences of method. Now, a 
method is just what we are looking for at present in 
order to enable us to pass from the study of the rules of 

1 Kant, Metaphysics of Ethics t pp. 9-10 and 104-113. 


games to the analysis of moral realities imposed upon 
the cMld by the adult. It is only from the point of view 
of the right method to adopt that we shall here shortly 
touch upon the vexed question of the individual -and 

One way of af tacking the problem is to analyse and 
explain the rules objectively, taking account of their 
connection with social groups defined by their morphology. 
This is the method which Durkheim used, and no one 
would think of denying his contribution to the subject 
of the evolution of moral realities. The mere fact of 
individuals living in groups is sufficient to give rise to 
new features of obligation and regularity in their lives. 
The pressure of the group upon the individual would 
thus explain the appearance of this sui generis feeling 
which we call respect and which is the source of all 
religion and morality. For the group could not impose 
itself upon the individual without surrounding itself with 
a halo of sanctity and without arousing in the individual 
the feeling of moral obligation. A rule is therefore nothing 
but the condition for the existence of a social group ; 
and if to the individual conscience rules seem to be 
charged with obligation, this is because communal life 
alters the very structure of consciousness by inculcating 
into it the feeling of respect. 

It is a striking fact, in this connection, that even such 
ephemeral groupings as those formed by children's 
societies or created primarily for the purpose of play 
have their rules and that these rules command the respect 
of individual minds. It is also curious to note how 
stable these rules remain in their main features and in 
their spirit throughout successive generations, and to 
what degree of elaboration and stylizatioa' they attain. 

But, as we have shown above, rules, although their 
.content continues to be the same, do not remain identical 
throughout the child's social development from the point 
of view of the kind of respect connected with them. 

For very young children, a rule is a sacred reality because 


it Is traditional ; for the older ones it depends upon mutual 
agreement. Heteronomy and autonomy are the two 
poles of this evolution. Does Durkheim's method enable 
us to explain these facts ? 

No one has felt more deeply than Durkheim nor sub- 
mitted to a more searching analysis the development and 
disappearance of obligatory conformity. In societies of 
a segmented type conformity is at its maximum : each 
social unit is a closed system, all the individuals are 
identical with each other except in the matter of age, 
and tradition leans with its full weight on the spirit 
of each. But as a society increases in size and density 
the barriers between its clans are broken down, local 
conformities are wiped out as a result of this fusion, and 
individuals can escape from their own people's super- 
vision. 'And above all, the division of labour which 
comes as the necessary result of this increasing density 
differentiates the individuals from one another psycho- 
logically and gives rise to individualism and to the 
formation of personalities in the true sense. Individual 
heteronomy and autonomy would thus seem to be in 
direct correlation with the morphology and the functioning 
of the group as a whole. 

Now, does this analysis apply to our children's societies ? 
In many respects, undoubtedly, it does. There is certainly 
a resemblance between segmented or mechanical solidarity 
and the societies formed by children of 5 to 8. As in the 
organized clan so in these groups, temporarily formed 
and isolated in relation to each other, the individual 
does not count. Social life and individual life are one. 
Suggestion and imitation are all-powerful. All individuals 
are alike except for differences of prestige and age. The 
traditional rule is coercive and conformity is demanded 
of an. 

As to the gradual disappearance of conformity as the 

child grows older, this too we could explain by some of 

the factors defined by Durkheim. To the increasing size 

and density of social groups and to the ensuing liberation 



of the Individual we can compare the fact that our 
children, as they grow older, take part in an ever-increasing 
number of local traditions. The marble player of 10 or 12 
will discover, for example, that there are other usages in 
existence besides those to which he is accustomed ; he 
will make friends with children from other schools who 
will free him from his narrow conformity, and in this 
way a fusion wiH take place between clans which up till 
then had been more or less isolated. At the same time, 
the growing child detaches himself more and more from 
his family circle, and since at first he assimilates games 
to the duties laid down for him by adults, the more he 
escapes from family conformity, the greater change will 
his consciousness of rules undergo. 

If, however, we are able to compare all these facts to 
the growth of societies in size and density, we can do so 
only from the point of view of the gradual diminution 
of the supervision exercised over individuals. In other 
words, the outstanding fact in the evolution of game 
rules is that the child is less and less dominated by the 
" older ones ". There is little or no progressive division 
of labour among children ; such differentiations as arise 
are psychological and not economic or political If, 
therefore, children's societies do, in a sense, develop from 
the segmented to the more highly organized type, and 
if there is a correlative evolution from conformity to 
individualistic cooperation, or from heteronoiny to auto- 
nomy, this process, though we may describe it in the 
objective terms of sociology, must be attributed first and 
foremost to the morphology and activity of the various 
age classes of the population. 

In other words, the main factor in the obligatory 
conformity of very young children is nothing but respect 
for age respect for older children, and, above all, respect 
for adults. And if, at a given moment, cooperation 
takes the place of constraint, or autonomy that of con- 
formity, it is because the child, as he grows older, becomes 
progressively free from adult supervision. This came out 


very clearly In the game of marbles. Children of n to 
13 have no others above them in this game, since it is 
one that is only played in the lower school But apart 
from this, the boy begins at this age to feel himself more 
and more on the same level as adolescents and to free 
himself inwardly from adult constraint. As a result, his 
moral consciousness undergoes the alterations we have 
outlined above. There can be no doubt that this pheno- 
menon is peculiar to our civilization and therefore falls 
under the Durkheimian scheme. In our societies the 
child of 13 escapes from the family circle and comes in 
contact with an ever-increasing number of social circles 
which widen Ms mental outlook. Whereas in so-called 
primitive communities, adolescence is the age of initiation, 
therefore of the strongest moral constraint, and the 
individual, as he grows older, becomes more and more 
dependent. But keeping in mind only our societies of 
children, we see that cooperation constitutes the most 
deep-lying social phenomenon, and that which has the 
surest psychological foundations. As soon as the individual 
escapes from the domination of age, he tends towards 
cooperation as the normal form of social equilibrium. 

In short, if, putting other considerations aside for the 
moment, we seek only to find a working hypothesis, the 
methodological difficulty of Durkheimism seems to be the 
following with regard to the different kinds of respect. 
Durkheim argues as though differences from one age or 
from one generation to another were of no account. 
He assumes homogeneous individuals and tries to find 
out what repercussion different modes of grouping would 
have upon their minds. All that he gets at in this way is 
profoundly true, but it is incomplete. We have only to 
make the impossible supposition of a society where every- 
one would be of the same age, of a society formed by a 
single generation indefinitely prolonged, to realize the 
immense significance attaching to age relations and 
especially to the relations between adults and children. 
Would such a society ever have known anything of 


obligatory conformity ? Would It be acquainted with 
religion or at any rate with the religions that taught 
transcendence? Would unilateral respect with al its 
repercussions upon the moral consciousness be observed 
in such a group as this ? We only wish to ask these 
questions. Whichever way they are answered, there can 
be no doubt that cooperation and social constraint deserve 
to be far more sharply contrasted than they usually are, 
the latter being perhaps nothing more than the pressure 
of one generation upon the other, whereas the former 
constitutes the deepest and most important social relation 
that can go to the development of the norms of reason. 

This influence exercised by age brings us to the second 
possible view of the psychology of rules, we mean that 
held by M. Bovet Theoretically, and in his method, 
M. Bovet recognizes only individuals. Only, instead of 
becoming involved, as others have been in a barren 
discussion on the limits of what is social and what is 
individual, M. Bovet admits that respect, the feeling of 
obligation, and the making of rules presuppose the inter- 
action of at least two individuals. On this point his 
method is parallel to Durkheim's and in no way opposed 
to it. For the real conflict lies between those who want 
to explain the moral consciousness by means of purely 
individual processes (habit, biological adaptation, etc.) 
and those who admit the necessity for an inter-individual 
factor. Once grant that two individuals at least must 
be taken into account if a moral reality is to develop, 
then it matters not whether you describe the facts 
objectively, as Durkheim did, or at least tried to do, or 
whether you describe them in terms of consciousness, 1 
How, asks M. Bovet, does the sense of duty appear? 
Two conditions, he says, are necessary, and their con- 
junction sufficient. i The individual must receive a 
command from another individual ; the obligatory rule 

1 See R. Lacombe's conclusive remarks, La Mtthode sociologique de 
Durhheim. Also d'Essertier, Psychologie et Sociologist Paris, Alcan, and 
many other contributions to the subject. 


is therefore psychologically different from the individual 
habit or from what we have called the motor rule. 2 The 
individual receiving the command must accept it, i.e. 
must respect the person from whom it came. M. Bovet 
differs on this point from Kant, since he regards respect 
as a feeling directed to persons and not to the rule as 
such. It is not the obligatory character of the rule laid 
down by an individual that makes us respect this indivi- 
dual, it is the respect we feel for the individual that 
makes us regard as obligatory the rule that he lays down. 
The appearance of the sense of duty in a child thus admits 
of the simplest explanation, namely that he receives 
commands from older children (in play) and from adults 
(in life), and that he respects older children and parents. 

It will be seen that our results completely confirm 
this view of the matter. Before the intervention of 
adults or of older children there are in the child's conduct 
certain rules that we have called motor rules. But they 
are not imperative, they do not constitute duties but 
only spontaneous regularities of behaviour. From the 
moment, however, that the child has received from his 
parents a system of commands, rules and, in general, 
the world order itself seem to him to be morally necessary- 
In this way, as soon as the little child encounters the 
example of older children at marbles, he accepts these 
suggestions and regards the new rules discovered in this 
way as sacred and obligatory. 

But the problem which faces us and which M. Bovet 
has himself clearly formulated and discussed is how this 
"morality of duty will allow for the appearance of the 
morality of goodness. 

The problem is two-fold. In the first place, the primitive 
"consciousness of duty is essentially heteronomous, since 
duty is nothing more than the acceptance of commands 
received from without. How then, asks M. Bovet, will 
~the child come to distinguish a " good " from a " bad " 
respect, and, after having accepted without distinction 
everything that was laid down for him by his environment, 


how will he leam to make Ms choice and to establish 
a hierarchy of values ? In language which exactly recalls 
that in which Durkheim describes the effect of increasing 
social density on the minds of the individuals, M. Bovet 
points here to the effect of conflicting influences and 
even of contradictory commands : the child pulled in 
several directions at once is forced to appeal to his reason 
in order to bring unity into the moral material. Already 
we have autonomy, but since reason does not create new 
duties and can only choose from among the orders 
received, this autonomy is still only relative. In the 
second place, alongside of the sense of duty we must, 
according to M. Bovet, distinguish a sense of goodness, 
a consciousness of something attractive and not merely 
obligatory, a consciousness that is fully autonomous. In 
contrast to Durkheim who, while he fully recognized this 
dualism of duty and good nevertheless tried to trace them 
both to the same efficient cause, viz. pressure of the group, 
M. Bovet leaves the question open, and does so in- 

It is at this point, so- it seems to us, that the part 
played by mutual respect comes in. Without going outside 
M. Bovet's fertile hypothesis, according to which all the 
moral sentiments are rooted in the respect felt by indivi- 
duals for each other, we can, nevertheless, distinguish 
different types of respect. It seems to us an undeniable 
fact that in the course of the child's mental development, 
unilateral respect or the respect felt by the small for the 
great plays an essential part : it is what makes the child 
accept ail the commands transmitted to him by his 
parents and is thus the great factor of continuity between 
different generations. But it seems to us no less undeniable, 
both in view of the results we have so far obtained and 
of the facts we shall examine in the rest of the book, 
that as the child grows in years the nature of his respect 
changes. In so far as individuals decide questions on an 
equal footing no matter whether subjectively or ob- 
jectively the pressure they exercise upon each other 


becomes collateral. And the interventions of reason, so 
rightly noted by M. Bovet, for the purpose of explaining 
the autonomy now acquired by morality, are precisely 
the outcome of this progressive cooperation. Our earlier 
studies led us to the conclusion that the norms of reason, 
and in particular the important norm of reciprocity, the 
source of the logic of relations, can only develop in and 
through cooperation. Whether cooperation is an effect 
or a cause of reason, or both, reason requires cooperation 
in so far as being rational consists in " situating oneself " 
so as to submit the individual to the universal. Mutual 
respect therefore appears to us as the necessary condition 
of autonomy under its double aspect, intellectual and 
moral. From the intellectual point of view, it frees the 
child from the opinions that have been imposed upon him 
while it favours inner consistency and reciprocal control. 
From the moral point of view, it replaces the norms of 
authority by that norm immanent in action and in con- 
sciousness themselves, the norm of reciprocity in sympathy . 
In short, whether one takes up the point of view of 
Durkheim or of M. Bovet, it is necessary, in order to 
grasp the situation, to take account of two groups of 
social and moral facts constraint and unilateral respect 
on the one hand, cooperation and mutual respect on the 
other. Such is the guiding hypothesis which will serve 
us in the sequel and which will lead us in examining 
the moral judgments of children to dissociate from one 
another two systems of totally different origin. Whether 
we describe the facts in the terms of social morphology 
or from the point of view of consciousness (and the two 
languages are, we repeat, parallel and not contradictory) 
it is impossible to reduce the effects of cooperation to 
those of constraint and unilateral respect. 



WE have had occasion to see during our analysis of the 
rales of a game that the child begins by regarding these 
rules not only as obligatory, but also as inviolable and 
requiring to be kept to literally. We also showed that 
this attitude was the result of the constraint exercised 
by the older children on the younger and of the pressure 
of adults themselves, rules being thus identified with duties 
properly so called. 

It is this problem of unilateral or one-sided respect, or 
of the effects of moral constraint, that we shall now 
approach through a more direct study of the child's 
conception of his duties and of moral values in general. 
But the subject is vast ; and we shall try to limit the 
range of enquiry as much as possible. We shall therefore 
confine ourselves to an aspect of the question which has 
perhaps received less attention than others the moral 
judgment itself. We were able before to observe con- 
cun"ently external and internal facts, to analyse both 
practice and consciousness of rules. But now, in view 
of the enormously greater technical difficulties which 
attend the study of the relations between children and 
adults, we shall have to limit ourselves to the consciousness 
of rules, and even to the most crystallized and least living 
part of this consciousness we mean what may be called 
the theoretical moral judgment as opposed to that which 
occurs in actual experience. But we are able. to confine 
ourselves to this special problem because numerous works 

1 In collaboration with M. N. Maso. 


have already told us all about the child's practice of 
moral rales and the conflicts that take place in Ms 
mind. A particularly large amount of work has been 
done, for example, on the subject of lying. Research of 
this kind is therefore the equivalent to the descriptions 
we have given of the practice of rules in the sphere of 
play, and it is therefore quite natural that we should 
confine ourselves to the study of children's judgments on 
such matters, judgments about lying, about truthful- 
ness, etc. 

In thus comparing the moral judgments of the child 
with what we know of his behaviour in the corresponding 
spheres of action, we shall endeavour to show that, as 
we were led to conclude in the case of game rales, the 
earliest forms assumed by a child's sense of duty are 
essentially heteronomous forms. We shall return, in this 
connection, to our hypotheses concerning the relations of 
heteronomy and egocentrism. Heteronomy, as we saw, 
was in no way sufficient to produce a mental change, 
and constraint and egocentrism were good bedfellows. 
This is more or less the same result as we shall find in 
studying the effects of adult constraint. Finally, we saw 
that cooperation was necessary for the conquest of moral 
autonomy. Now, such a hypothesis can be proved only by 
a close analysis of the way in which moral rules are at a 
given moment assimilated and freely adopted by the child. 

In this chapter we shall study primarily the effects of 
moral constraint, though we shall also establish some of 
the landmarks for the outline of cooperation which will 
be given later on. Now, moral constraint is closely 
akin to intellectual constraint, and the strictly literal 
character which the child tends to ascribe to rules received 
from without bears, as we shall see, a close resemblance 
to the attitudes he adopts with regard to language and 
the intellectual realities imposed upon him by the adult. 
We can make use of this analogy to fix our nomenclature 
and shall speak of moral realism to designate on the plane 
of judgments of value what corresponds to " nominal 


realism " and even verbalism or conceptual realism on 
the plane of theoretical reasoning. Not only this, but 
just as realism In general (in the sense in which we have 
used the word in our previous books, see C.W., first part) 
results both from a confusion between subjective and 
objective (hence from egocentrism) and from the in- 
tellectual constraint of the adult, so also does moral 
realism result from the intersection of these two kinds 
of causes. 

We shall therefore call moral realism the tendency 
which the child has to regard duty and the value attaching 
to it as self-subsistent and independent of the mind, as 
imposing itself regardless of the circumstances in which 
the individual may find himself. 

Moral realism thus possesses at least three features. In 
the first place, duty, as viewed by moral realism, is essenti- 
ally heteronomous. Any act that shows obedience to a rule 
or even to an adult, regardless of what he may command, 
is good ; any act that does not conform to rules is bad* 
A rule is therefore not in any way something elaborated, 
or even judged and interpreted by the mind; it is given as 
such, ready made and external to the mind. It is also 
conceived of as revealed by the adult and imposed by him. 
The good, therefore, is rigidly defined by obedience. 

In the second place, moral realism demands that the 
letter rather than the spirit of the law shall be observed. 
This feature derives from the first. Yet it would be possible 
to imagine an ethic of heteronomy based on the spirit of 
the rules and not on their most hard and fast contents. 
Such an attitude would already have ceased to be realist ; 
it would tend towards rationality and inwardness. But 
at the very outset of the moral evolution of the child, 
adult constraint produces, on the contrary, a sort of 
literal realism of which we shall see many examples 
later on. 

In the third place, moral realism induces an objective 
conception of responsibility. We can even use this as a 
criterion of realism, for such an attitude towards re- 


spOBsIbMty is easier to detect than the two that precede 
it. For since he takes rules literally and thinks of good 
only in terms of obedience, the child will at first evaluate 
acts not in accordance with the motive that has prompted 
them but in terms of their exact conformity with established 
rules. Hence this objective responsibility of which we shall 
see the clearest manifestations in the moral judgment of 
the child. 

I. THE METHOD. Before proceeding to the analysis of 
the facts, it will be as well to discuss in a few words the 
method we propose to adopt. The only good method in 
the study of moral facts is surely to observe as closely as 
possible the greatest possible number of individuals. 
Difficult children, whom parents and teachers send or 
ought to send up for psycho-therapeutic treatment, supply 
the richest material for analysis. In addition to this, 
education in the home constantly gives rise to the most 
perplexing problems. Now, the removal of these problems 
cannot always, unfortunately for the children, depend solely 
upon the " common sense " of the parents, and the educa- 
tional technique necessary for their solution is, in some 
respects, the best instrument of analysis at the disposal 
of the psychologist. We shall therefore do our utmost 
in the sequel to give as valid only such results as do 
not contradict observation in family life. 

Only, here again, as in the case of intellectual notions, 
while pure observation is the only sure method, it allows 
for the acquisition of no more than a small number of 
fragmentary facts. And we therefore consider that it 
must be completed by questioning children at school. 
We shall speak of these interrogatories now, whilst we 
may have to postpone until a later date the publication 
of the observations we have been able to make on our 
own children. If, however, questioning in the intellectual 
field is relatively easy, in spite of the many difficulties 
of method which it raises, in the moral sphere it can only 
be, as it were, about reality once removed. You can 


make a child reason about a problem of physics or logic. 
That brings you into contact, not indeed with spontaneous 
^ thought, but at least with thought in action. But you 
cannot make a child act in a laboratory In order to dissect 
Ms moral conduct. A moral problem presented to the 
child is far further removed from Ms moral practice than 
is an intellectual problem from his logical practice. It is 
only in the domain of games if there that the methods 
of the laboratory will enable us to analyse a reality in 
the making. As to the moral rules which the child 
receives from the adult, no direct investigation is to be 
thought of by interrogation. Let us therefore make the 
best of it and try to examine, not the act, but simply 
the judgment of moral value. In other words, let us 
analyse, not the child's actual decisions nor even his 
memory of his actions, but the way he evaluates a given 
piece of conduct. 

Here, moreover, a fresh difficulty raises its head. We 
shall not be able to make the child realize concretely the 
types of behaviour that we submit to him for judgment, 
as we could in handing him a game of marbles or a 
mechanism of any sort. We shall only be able to describe 
them by means of a story, obviously a very indirect 
method. To ask a child to say what he thinks about 
actions that are merely told to him can this have the 
least connection with child morality ? On the one hand, 
it may be that what the child thinks about morality has 
no precise connection with what he does and feels in his 
concrete experience. Thus the interrogatory of children 
of 5 to 7 about marbles revealed the strangest discrepancy 
between actual practice of the rules and reflection about 
them. On the other hand, it may also be that what the 
child actually understands of the stories suggested to him 
bears no relation to what he would think if he were to 
witness these scenes himself. 

We must not attempt to solve these difficulties of 
method by means of any a priori considerations. We 
wish only to draw attention to them and to the theoretical 


Interest of the problems which they raise. For we are 
faced with purely general questions on the relations that 
hold between verbal judgment and the practical applica- 
tion of thought, whether intellectual or moral. It is 
quite true that research on intelligence is easier than on 
morality; but this holds only of the functioning of 
thought, not of its content. When, in order to get at 
this content, one is obliged, as we have been in the past, 
to question the child about his own beliefs, the problem 
is the same. The question may therefore be formulated 
as follows. Does verbal thought, i.e. thought that works 
jjpon ideas evoked by language and not upon objects 
perceived in the course of action, does verbal thought 
consist in the conscious realization 1 (allowing of course 
for various systematic distortions) of truly spontaneous 
thought, or does it sustain with the latter no relations 
whatever ? Be the answer what it may, this question is 
one of fundamental importance in human psychology. 
Is man merely a maker of phrases that have no relation 
to his real actions, or is the need to formulate part of his 
very being ? The question strikes deep, and to solve it 
we must, amongst other things, study it in the child. 
In the child as in ourselves there is a layer of purely 
.verbal thought superposed, as it were, over his active 
thought. It is not only during the interrogatories that 
he invents stories. He is telling them to himself all 
the time, and it is relatively easy to prove that the 
stories invented in psychological experiments are roughly 
analogous to those that arise spontaneously. (We were 
able to show, for example, that the results obtained by 
questioning the child on the various aspects of Ms con- 
ception of the world corresponded in the main with what 
was revealed by direct observation and .by the analysis 
of "whys" in particular.) But the problem remains. 
In what relation does the verbal thought of the child 
stand to Ms active and concrete thought ? 
The problem is of special interest in the sphere of 

1 See footnote on p. 56. 


morality, and the difficulties of the method we are going 
to use must be subjected to systematic scrutiny. The 
object of this scrutiny will not be the justification or 
condemnation of our method (any method that leads to 
constant results is interesting, and only the meaning of 
the results is a matter for discussion) but to help towards 
a more definite statement of the problem of "moral 
theories ". 

To begin with the adult. There are, on the one hand, 
authors who deem it indispensable for the mind to codify 
its norms, or at any rate to reflect upon the nature of 
moral action. Such people therefore accept, with or 
without discussion, this postulate of an existing relation 
between moral reflection and moral practice saying 
either that the latter springs from the former, or that 
reflection is the conscious realization of action or action 
coming into consciousness. There are, on the other hand, 
individuals whose personal conduct, incidentally, may be 
beyond criticism, but who do not believe in " morality ". 
Kant and Durkheim. are typical representatives of the 
first tendency, Pareto is the most typical of living 
authors of the second. 1 According to him, only actions 
"exist, some of which are logical, others non-logical, i.e. 
instinctive, or coloured with affectivity. Added to this 
and on a completely different plane there is a sort of 
rambling chatter, whose function is to reinforce action 
but whose contents may be devoid of any intelligible 
meaning. This chatter multiform and arbitrary " deri- 
vktions " founded on the affective residues of non-logical 
actions this chatter is what constitutes our ethical 
theories ! 

The point, then, that we have to settle is whether the 
things that children say to us constitute, as compared tp 
their real conduct, a conscious realization or a " deriv'a- 
tion ", reflection (in the etymological sense of the word) or 
psittacism. . . . We do not claim to have solved the 
problem completely. Only direct observation can settle it. 

1 Tvaiti de Sociologie Ginirale, 2 Vol., Payot. 


But to enable us to give the casting vote to observation, 
we shall first have to find out what are the child's verbal 
ideas on morality. And this is why we consider our 
researches to be useful, whatever may be their ultimate 
result. Moreover, the study of rales which we undertook 
in the last chapter has already suppled us with the most 
precious indication. Broadly speaking, we found there 
to be in this domain a certain correspondence (not simple 
but yet quite definable) between children's judgments 
about rules and their practice of these same rules. Let us 
therefore carry our analysis of the problem a stage further. 

What, in the first place, are the relations between 
"judgment of value and the moral act itself? Here is 
a child who declares it to be perfectly legitimate to 
tell Ms father about Ms brother's misdeeds. Another 
child answers that even if the father asks, it is " horrid " 1 
(Fr., vilain) to tell tales : it is better to " spin a yam " 
(Fr. dire des blagues) than to let a brother be punished. 
The problem is to know whether in practice these two 
children would really have considered valid the two 
courses of action wMch they recommend verbally. 

We must be on our guard here against a certain 
ambiguity. Some experimenters have tried to measure 
the moral value of a child by testing of Ms moral judg- 
ment. Mile Descoeudres, for example, holds the view 
that a child who pronounces correctly on the values of 
actions he is told about, is, on the whole, better than one 
whose moral judgment is less acute, 2 This may be, but it 
is also conceivable that intelligence alone might suffice to 
sharpen the child's evaluation of conduct without neces- 

1 While in most of the interrogatories we have translated vilain by 
" naughty *', the reader should note that the English word has an 

."exclusively authoritative ring which the French has not. Children 
can only be naughty in reference to grown-ups. Indeed, the word is 

_so powerful a weapon in the hands of adult constraint in this country 
that its use in any verbal experiments made on English children would 
probably give appreciably different results from those based on the word 
vilain, [Trans.] 

f A. Descoeudres, "Sur le Jugement moral/' L' Jntermtdiaire des, 
Educateurs, II, p. 54 (1914). 


sarily induing Mm to do good actions. In this case an 
intelligent scamp would perhaps give better answers than 
a slow-witted but really good-hearted little boy. Besides, 
how is the psychologist to classify the moral worth of 
children even by the ordinary common-sense standards ? 
Such a classification, possible in extreme cases, would run 
the risk of inaccuracy in normal cases, which are precisely 
those where we want to know whether testing moral 
Judgment will help us to know the child. 

But apart from this question, which does not really 
interest us here, it may well be asked whether the judgment 
of value given by the child during an interrogatory is the 
sSkme as he would give in practice, independently of the 
actual decision which he would take. A given child, for 
example, will tell us during the interrogation that the lie 
a is worse than the lie b. Now, whether he tells lies him- 
self or not, whether, that is, he is or is not what he calls 
" good ", we take the liberty to wonder whether, in 
action, he will still consider lie a worse than lie b. What 
we are after is not how the child puts his moral creed into 
practice (we saw in connection with the game of marbles 
that a mystical respect for rules can go hand in hand with 
a purely egocentric application of them) but how he 
judges of good and evil in the performance of his own 
actions. It is from this point of view only that we set 
ourselves the problem of discovering whether the judg- 
ments of value given in the interrogatories do or do 
not correspond to the genuine evaluations of moral 

Now, it may be that there is correlation between verbal 
or theoretical judgment and the concrete evaluations that 
operate in action (independently of whether these evalua- 
tions are followed up by real decisions). We have often 
noted that in the intellectual field the child's verbal 
thinking consists of a progressive coming into conscious- 
ness, or conscious realization of schemas that have been 
built up by action. In such cases verbal thought simply 
lags behind concrete thought, since the former has to 


reconstruct symbolically and on a new plane operations 
that have already taken place on the preceding level. 
Old difficulties, which have been overcome on the plane 
of action will therefore reappear or merely survive on the 
verbal plane. There is a time-lag between the concrete 
phases and the verbal phases of one and the same process. 
It may therefore very well be that in the moral sphere 
there is simply a time-lag between the child's concrete 
evaluations and his theoretical judgment of value, the 
latter being an adequate and progressive conscious realiza- 
tion of the former. We shall meet with children who, 
for example, take no account of intentions in appraising 
actions on the verbal plane (objective responsibility), but 
who, when asked for personal experiences, show that they 
.take full account of the intentions that come into play. 
It may be that in such cases the theoretical simply lags 
behind the practical moral judgment and shows in an 
adequate manner a stage that has been superseded on the 
plane of action. 

But there may also be no connection whatever between 
the two. On this view, the child's moral theories would 
be mere .chatter, unrelated to his concrete evaluations. 
Further and the eventuality is still more important in the 
moral than in the intellectual sphere it may be for the 
benefit of the adult rather than for his own use that the 
child gives his answers. Let there be no mistake on this 
point : it is quite certain that in the great majority of 
cases the child is perfectly sincere during the experiment. 
Only, he is quite likely to think that what is expected of 
him is a moral lecture rather than an original reflection. 
We have talked with children of 10, for example, who 
defended the moral value of " telling tales ", but who 
made a volte face as soon as they saw we were not con- 
vinced. Thus their real thought was masked and hidden 
even from their own eyes by the momentary desire to 
pronounce moral precepts pleasing to the adult. True, 
,'only the older ones reacted in this way. But does not this 
show that the little ones do not dissociate their own 


thought from what they hear constantly being said by 
their parents and teachers ? Is it not the case that the 
verbal thought of the child up to 10-11 is simply a 
repetition or a distortion of adult thought, bearing no 
relation to the real moral evaluations which the child 
practises in Ms own life ? 

To settle this point we can turn to what we learned in 
our enquiry into the game of marbles. On the one hand, 
we were able to see how the children put the rules into 
practice and how they evaluated their duties as players 
in the midst of the game itself. On the other hand, we 
succeeded in collecting on the subject of these same rules 
certain moral theories obviously made up on the spot and 
therefore grounded on purely theoretical moral judgments. 
Now there was, we repeat, between the action and the 
theory of the child a correspondence that was, if not 
simple, at least definable. To the egocentric practice of 
rales which goes hand in hand with a feeling of respect 
for elder or adult, there corresponds a theoretical judg- 
ment which" turns a rule into something mystical and 
transcendental. In this first case, theoretical judgment 
does not correspond to action itself but to the judgments 
that accompany action. But this is quite natural, since 
egocentrism is unconscious, and only the respect to which 
the child believes that he is submitting himself is con- 
scious. To the rational practice of rules, which goes hand 
in hand with mutual respect, there corresponds a theoreti- 
cal judgment which attributes to rules a purely autonomous 
character. Thus in the sphere of play at least, theoretical 
judgment corresponds to practical judgment. This does 
not mean that theoretical judgment interprets the child's 
real action but that in the main it corresponds to the 
judgments pronounced by the child in the course of his 
action. We can at the most admit that verbal judgment 
lags behind effective judgment : the idea of autonomy 
appears in the child about a year later than cooperative 
behaviour and the practical consciousness of autonomy. 

With regard to the domains we are now approaching 


(lying, justice, etc.) we may therefore advance the 
hypothesis that the verbal and theoretical judgment of the 
child corresponds, broadly speaking, with the concrete 
and practical judgments which the child may have made on 
the occasion of Ms own actions during the years preceding 
the interrogatory. There can be no doubt that verbal 
thought lags behind active thought, but it does not seem 
to us to be unrelated to the past stages of active thought. 
The future will show whether this hypothesis is too bold. 
In any case, verbal thought, whether moral or intellectual, 
deserves the closest study. Nor is it peculiar to the child. 
In the adult, as the work of Pareto sufficiently shows, it 
plays a considerable part in the mechanism of social life. 

Finally, allowance must be made for the fact that the 
verbal evaluations made by our children are not of actions 
of which they have been authors or witnesses, but of 
stories which are told to them. The child's evaluation 
will therefore be, as it were, verbal to the second degree. 
The psychologist Fernald * has tried to obviate this dis- 
advantage by the following device. He tells the children 
several stories and then simply asks them to classify them* 
Mile Descoeudres, applying this method, submits, for 
example, five lies to children, who are then required to 
classify them in order of gravity. This, roughly, is also 
the procedure that we shall follow, though we shall of 
course not deny ourselves the right, once the classifica- 
tion has been made, to converse freely with the children 
so as to get at the reasons for their evaluations. 

But the greatest caution must be exercised in order to 
avoid needless complications. For instance, it does not 
seem to us possible to tell the children more than two 
stories at a time. If the subject is confronted with a 
series, the classification will call for an intellectual effort 
that has nothing to do with moral evaluation : he will 
forget three stories out of five, and will compare any two 
at random, which gives results of no particular interest. 
Further, after using the usual stories, we soon realized that 

1 Fernald, American Journal of Insanity* April 1912. 


their style placed them far beyond the child's complete 
comprehension. In psychology one must speak to children 
in their own language, otherwise the experiment resolves 
itself into a trial of intelligence or of verbal understanding. 

But even when we have taken all these precautions, one 
problem still remains : if the child had witnessed the 
scenes we describe to him, would he judge them in the 
same manner ? We think not. In real life the child is in 
the presence, not of isolated acts, but of personalities that 
attract or repel him as a global whole. He grasps 
people's intentions by direct intuition and cannot there- 
fore abstract from them. He aEows, more or less justly, for 
aggravating and attenuating circumstances. This is why 
the stories told by the children themselves often give rise 
to different evaluations from those suggested by the 
experimenter's stories. Only, we repeat it, it may simply 
be the case that the evaluations obtained from the stories 
that were told to them lag .in time behind the direct 
evaluations of daily life. 

In conclusion, the results of our method do not seem to 
us devoid of interest. For they are relatively constant 
and, above all, they evolve with a certain regularity 
according to age. All that we have said before about the 
criteria of good clinical interrogatories (C.W., Introd.) 
applies here. And, in addition, it is our belief that in 
everyday life, as in the course of the interrogatory, the 
child must often be faced not only with concrete actions 
but also with accounts and verbal appraisals of actions. 
It is therefore important to know what is his attitude in 
such circumstances. In short, here as always, the way of 
really tackling the problem is not to accept and record 
the results of the experiment, but to know how to place 
them in regard to the child's real life taken as a whole. 
And this cannot be done at the outset of our enquiry into 
this most difficult field of research. 

STEALING. We noted, in connection with the rules of a 


game, that the child seems to go through a stage when 
rules constitute an obligatory and untouchable reality. We 
must now see how far this moral realism goes; and in 
particular whether adult constraint, which is probably its 
cause, is sufficient to give rise to the phenomenon of objective 
responsibility. For all that we have been saying about 
the difficulties of interpretation in the study of the moral 
judgments of children need not put a stop to our enquiry 
in this matter. It is immaterial whether the objective 
responsibility of which we are about to give examples is 
connected with the whole of the child's life or only with 
the most external and verbal aspects of his moral thought. 
The problem still remains as to where this responsibility 
comes from and why it develops. 

The questions put to the children on this point are 
those whose results we shall study first, but they were 
actually the last that we thought of. We began, by way 
of introduction, with the problem of judgments relating 
to telling lies. In making this analysis, of which we shall 
speak in the following sections, we immediately noticed that 
the younger children often measured the gravity of a lie not 
in terms of the motives which dictated it, but in terms of 
the falseness of its statements. It was in order to verify the 
existence and the generality of this tendency to objective 
responsibility that we devised the following questions. 

The first set of questions deals with the consequences 
of clumsiness. Clumsiness plays, however unjustly, an 
enormously important part in a child's life, as he comes 
into conflict with his adult surrounding. At every 
moment, the child arouses the anger of those around him 
by breaking, soiling, or spoiling some object or other. 
Most of the time such anger is unjustifiable, but the child 
is naturally led to attach a meaning to it. On other 
occasions, his clumsiness is more or less due to carelessness 
or disobedience, and an idea of some mysterious and 
immanent justice comes to be grafted on to the emotions 
experienced at the time. We therefore tried to make the 
children compare the stories of two kinds of clumsiness, 


one, entirely fortuitous or even the result of a well-in- 
tentioned act, but involving considerable material damage, 
the other, negligible as regards the damage done but 
happening as the result of an ill-intentioned act. 
Here are the stories : 

I. A. A little boy who is called John is in his room. 
He is called to dinner. He goes into the dining room. 
But behind the door there was a chair, and on the chair 
there was a tray with fifteen cups on it. John couldn't 
have known that there was all this behind the door. He 

foes in, the door knocks against the tray, bang go the 
fteen cups and they all get broken 1 

B. Once there was a little boy whose name was 
Henry. One day when his mother was out he tried to get 
some jam out of the cupboard. He climbed up on to a 
chair and stretched out his arm. But the jam was too high 
up and he couldn't reach it and have any. But while he 
was trying to get it he knocked over a cup. The cup fell 
down and broke. 

II. A. There was a little boy called Julian. His father 
had gone out and Julian thought it would be fun to play 
with Ms father's ink-pot. First he played with the pen, 
and then he made a little blot on the table cloth. 

B. A little boy who was called Augustus once noticed 
that his father's ink-pot was empty. One day that Ms 
father was away he thought of filling the ink-pot so as to 
help Ms father, and so that he should find it full when he 
came home. But while he was opening the ink-bottle he 
made a big blot on the table cloth. ^ 

III. A. There was once a little girl who was called 
Marie. She wanted to give her mother a nice surprise, 
and cut out a piece of sewing for her. But she ^didn't 
know how to use the scissors properly and cut a big hole 
in her dress. 

B. A little girl called Margaret went and took her 
mother's scissors one day that her mother was out. She 
played with them for a bit. Then as she didn't know how 
to use them properly she made a little hole in her dress. 

When we have analysed the answers obtained by means 
of these pairs of stories, we shall study two problems 
relating to stealing. As our aim is for the moment to find 
out whether the cMld pays more attention to motive or 
to material results, we have confined ourselves to the 


comparison of selfishly motivated acts of stealing with 
those that are well-intentioned. 

IV. A. Alfred meets a little friend of Ms who is very 
poor. This friend tells Mm that he has had no dinner 
that day because there was notMng to eat in Ms home. 
Then Alfred goes into a baker's shop, and as he has no 
money, he waits till the baker's back is turned and steals 
a roll. Then he runs out and gives the roll to Ms friend. 

B. Heniiette goes into a shop. She sees 4 pretty 
piece of ribbon on a table and thinks to herself that it 
would look very nice on her dress. So wMle the shop 
lady's back is turned (while the shop lady is not looking), 
she steals the ribbon and runs away at once. 

V. A. Albertine had a little friend who kept a bird in a 
cage. Albertine thought the bird was very unhappy, and 
she was always asking her friend to let him out. But the 
friend wouldn't. So one day when her friend wasn't 
there, Albertine went and stole the bird. She let it fly 
away and hid the cage in the attic so that the bird should 
never be shut up in it again. 

B. Juliet stole some sweeties from her mother one 
day that her mother was not there, and she hid and ate 
them up. 

About each of these pairs of stories we ask two ques- 
tions : i Are these children equally guilty (or as the 
young Genevese say " la meme chose vilain " *) ? 2 
WMch of the two is the naughtiest, and why ? It goes 
without saying that each of these questions is the occasion 
for a conversation more or less elaborate according to the 
child's reaction. It is also as well to make the subjects 
repeat the stories before questioning them. The way the 
child reproduces the story is enough to show whether he 
has understood it. 

We obtained the following result. Up to the age of 10, 
two types of answer exist side by side. In one type 
actions are evaluated in terms of the material result and 
independently of motives ; according to the other type of 
answer motives alone are what counts. It may even 
happen that one and the same child judges sometimes one 

1 See foot-note, p in. 


way, sometimes the other. Besides, some stories point 
more definitely to objective responsibility than others. 
In detail, therefore, the material cannot be said to embody 
stages properly so called. Broadly speaking, however, it 
cannot be denied that the notion of objective responsibility 
diminishes as the child grows older. We did not come 
across a single definite case of it after the age of 10. In 
addition, by placing the answers obtained under 10 into two 
groups defiiied respectively by objective and by subjective 
responsibility (reckoning by answers given to each story 
and not by children, since each child is apt to vary from 
one story to another) we obtained 7 as the average age for 
objective responsibility, and 9 as the average age for 
subjective responsibility. Now, we were unable to ques- 
tion children under 6 with any profit because of the 
intellectual difficulties of comparison. The average of 7 
years therefore represents the youngest of the children. 
If the two attitudes simply represented individual types 
or types of family education, the two age averages ought 
to coincide. But since this is not so, there must be some 
degree of development present. We can at least venture 
to submit that even if the objective and the subjective 
conceptions of responsibility are not, properly speaking, 
features of two successive stages, they do at least define 
two distinct processes, one of which on the average 
precedes the other in the moral development of the child, 
although the two partially synchronize. 

Having made this point clear, let us now turn to the 
facts, beginning with the stories about clumsiness. Here 
are typical answers showing a purely objective notion of 



you understood these stories ? Yes. What did the first 
boy do ? He broke eleven cups. And the second one ? 
He broke a cup by moving roughly. Why did the first one 
break the cups ? Because the door knocked them. And 
the second ? He was clumsy. When he was getting the jam 
the cup fell down. Is one of the boys naughtier than the 


other ? The first is because he knocked over twelve cups. 
If you were the daddy, which one would you punish 
most ? The one who broke twelve cups. Why did he break 
them ? The door shut too hard and knocked them. He 
didn't do it on purpose. And why did the other boy break 
a cup ? He wanted to get the jam. He moved too far. The 
cup got broken. Why did he want to get the jam ? 
Because he was all alone. Because Ms mother wasn't there. 
Have you got a brother ? No, a little sister. Well, if it 
was you who had broken the twelve cups when you went 
into the room and your little sister who had broken one 
cup while she was trying to get the jam, which of you 
would be punished most severely ? Me, because I broke 
more than one cup. 13 

SCHMA (6) : " Have you understood the stories ? Let's 
hear you tell them.- A little child was called in to dinner. 
There were fifteen plates on a tray. He didn't know. He 
opens the door and he breaks the fifteen plates. That's very 
good. And now the second story ? There was a child. 
And then this child wanted to go and get some jam. He gets 
on to a chair, his arm catches on to a cup, and it gets broken. 
Are those children both naughty, or is one not so 
naughty as the other ? Both just as naughty. Would you 
punish them the same ? No. The one who broke fifteen 
plates. And would you punish the other one more, or 
less ? The first broke lots of things, the other one fewer. 
How would you punish them ? The one who broke the 
fifteen cups : two slaps. The other one, one slap." 

CONST (7) G. : " Tell me those two stories. There was 
a chair in the dining room with cups on it. A boy opens the 
door, and all the cups are broken, And now the other 
story ? A little boy wants to take some jam. He tried to 
take hold of a cup and it broke. II you were their mother, 
which one would you punish most severely ? The one who 
broke the cups. Is he the naughtiest ? Yes. Why did he 
break them ? Because he wanted to get into the room. 
And the other ? Because he wanted to take the jam. 
Let's pretend that you are the mummy. You have two 
little girls. One of them breaks fifteen cups as she is 
coming into the dining room, the other breaks one cup as 
she is trying to get some jam while you are not there. 
Which of them would you punish most severely ? The 
one who broke the fifteen cups. 1 ' But Const who is so 
decided about our stories goes on to tell us some personal 
reminiscences in which it is obviously subjective responsi- 


bility that is at work. " Have you ever broken anytMng ? 
A cup. How ? / wanted to wipe it, and I let it drop. 
What else have you broken ? Another time, "*a plate. 
How ? I took it to play with. Which was the naughtiest 
thing to do ? The plate, because I oughtn't to have taken it. 
And how about the cup ? That was less naughty 
because I wanted to wipe it. Which were you punished 
most for, for the cup or for the plate ? For the plate. 
Listen, I-am-goixig to tell you two more stories. A little 
girl was wiping the cups. She was putting them away, 
wiping them with the cloth, and she broke five cups. 
Another little girl is playing with some plates. She 
breaks a plate. Which of them is the naughtiest ? The 
one who broke the five cups." This shows that in the case 
of her own personal recollections (where, incidentally, the 
number of objects broken does not come in) subjective 
responsibility alone is taken into account. As soon as we 
go back to the stories, even basing them on the child's 
recollections, objective responsibility reappears in all its 
purity ! 


whose answers we have just been examining repeats 
correctly the story of the blot of ink : " A little boy sees 
that his father's ink-pot is empty. He takes the ink-bottle, 
but he is clumsy and makes a big blot. And the other one ? 
There was a boy who was always touching things. He 
takes the ink and makes a little blot. Are they both equally 
naughty or not ? No. Which is the most naughty ? 
The one who made the big blot. Why ? Because it was 
big. Why did he make a big blot ? To be helpful. And 
why did the other one make a little blot ? Because he 
was always touching things. He made a little blot. Then 
which of them is the naughtiest ? The one who made a 
big blot: 3 

GEO (6) also understands the stories and knows that the 
two children's intentions were quite different. But he 
regards as the naughtiest " the one who made the big blot. 
Why ? Becaiise that blot is bigger than the other one." 

III. THE STORY OF THE HOLES. GEO (6) is equally 
successful in understanding these two stories. " The first 
panted to help her mother and she made a big hole in her 
frock. The other one was playing and made a little hole. 
Is one of these little girls naughtier than the others ? 
The one who wanted to help her mother a little is the 
naughtiest because she made a big hole. She got scolded" 


CONST (7) G. repeats the stories as follows : " A 
girl wanted to make a handkerchief for her mother. She was 
clumsy, and made a big hole in her frock. And the other 
one ? There was a little girl who was always touching 
things. She took some scissors to play and made a little hole 
in her frock. Which of them is naughtiest IThe one who 
made the big hole. Why did she make this hole ? She 
wanted to give her mother a surprise. That's right. And 
the other one ? She took the scissors because she ^ was 
always touching things and made a little hole. That's right. 
Then which of the little girls was nicest ? , . . (hesita- 
tion). Say what you think. The one who made the little 
hole is the nicest. If you were the mother you would have 
seen everything they did. Which would you have punished 
most ? The one who made a big hole. And which one 
would you have punished least ? The one who made the 
little hole. And what would the one who made the big 
hole say when you punished her most ? She would say, I 
wanted to give a surprise. And the other one ? She was 
playing. Which one ought to be punished most ? The 
one who made the big hole. Let's pretend that it was you 
who made the big hole so as to give your mother a surprise. 
Your sister is playing and makes the little hole. Which 
ought to be punished most ? Me. Are you quite sure, 
or not quite sure? Quite sure. Have you ^ ever made 
holes ? Never. Is what I am asking you quite easy ? 
Yes. Are you quite sure you meant what you said ? 

These answers reveal the strength of the resistance 
offered to the counter-suggestions we attempted to make, 
and they also show what store the children set by material 
results, in spite of the fact that they have perfectly well 
understood the story and consequently the intentions of 
its characters, and what little account they take of the 
intentions which have indirectly caused these material 

Such facts as these taken by themselves of course prove 
nothing. Before speaking about objective responsibility, 
we must ask ourselves whether the child does not draw a 
distinction analogous to that which the adult makes in the 
case of ethics and of certain legal punishments. One can 
without any loss of honour be run in for having broken 


police regulations. One can be the object of a legal 
sentence devoid of any penal element (cf. Durkheim's 
restitutive and retributive punishment). In the same 
way, then, when a child pronounces a little girl to be 
" naughty " because she has made a big hole in her dress, 
although he knows that her intentions were not only 
innocent but admirable, does he not simply mean that she 
has damaged her parents materially and therefore deserves 
a purely legal punishment devoid of any moral signifi- 
cance ? 

The question arises in the same form in connection with 
stealing, as we shall see presently. But with regard to 
lying, since all question of material damage can be dis- 
regarded, we shall endeavour to prove that the child's 
judgments really do imply objective responsibility. An 
analogous conclusion may therefore be formulated con- 
cerning the present examples. Here preoccupation about 
material damage certainly outweighs any question of 
obedience or disobedience to rules. But this is a form of 
objective responsibility only in so far as the child fails to 
distinguish the element of civic responsibility, as it were, 
from the penal element. Now, on the verbal plane where 
we have taken up our stand it seems to us that this 
differentiation is one that hardly enters into the subject's 
mind. Responsibility is thus still held to be objective, 
even from the moral point of view. 

Before carrying our analysis any further and in order 
to place the previous attitudes in their true perspective, 
let us examine the answers that contradict those which 
we have just dealt with and which relate to the same 
pairs of stories : 

I. STORY OF THE BROKEN CUPS. Here, to begin with, is 
a rather exceptional case of a 6-year-old child. (Most of 
the children of 6 gave us answers which corresponded to 
the type of objective responsibility.) SCHMA (6J, G., 
forward intellectually and looking more like a girl of 8) 
begins by telling us that the two boys of the story are 
" equally naughty" and that they must be punished " Both 
just the same. 1 ' " Well, I think one of them is naughtier 


than the other. Which one do you think ? Both the same. 
Have you never broken anything ? No, I never ham. 
My brother has. What did he break ? A cup and a pail. 
How ? He wanted to fish. He broke half my pail, and 
then afterwards he broke it again on purpose to annoy me. 
Did he also break a cup ? He had wiped it and was putting 
it on the edge of the table and it fell. What day was he 
naughtiest, the day he broke the pail or the day he broke 
the cup ? The paiLWhy ?He broke my pail on pur- 
pose. And the cup ? He didn't do that on purpose. He 
put it right on the edge and it broke. And in the stories I 
told you, which boy is naughtiest, the one who broke the 
fifteen cups or the one who broke one cup ? The one who 
wanted to take the jam because he wanted to eat it." Thus 
ky appealing to her personal memories one sees that 
Schma can be led to judge according to subjective 

MOL (7) : " Which is naughtiest ? The second, the one 
who wanted to take the jam-pot, because he wanted to take 
something without asking. Did he catch it ? No. Was 
he the naughtiest all the same ? Yes. And the first ? 
It wasn't his fault. He didn't do it on purpose." 

CORM (9) : " Well, the one who broke them as he was 
coming isn't naughty, 'cos he didn't know there was any cups. 
The other one wanted to take the jam and caught his arm on 
a cup. Which one is the naughtiest? The one who 
wanted to take the jam. How many cups did he break ? 
One. And the other boy ? Fifteen. Which one would 
you punish most ? The boy who wanted to take the jam. 
He knew, he did it on purpose.' 9 

GROS (9) : " What did the first one do IHe broke 
fifteen cups as he was opening a door. And the second 
one ? He broke one cup as he was taking some jam. 
Which of these two silly things was naughtiest, do you 
think ? The one where 'he tried to take hold of a cup was 
[the silliest] because the other boy didn't see [that there were 
some cups behind the door]. He saw what he was doing. 
How many did he break ? One cup. And the other one ? 
Fifteen. Then which one would you punish most ? 
The one who broke one cup. Why ? He did it on purpose. 
If he hadn't taken the jam, it wouldn't have happened." 

Nuss (10) : The naughtiest is " the one who wanted to 
take the jam. Does it make any difference the other one 
having broken more cups ? No, because the one who broke 
fifteen cups didn't do it on purpose" 


II. STORY OF THE INK-STAINS. Sci (6) : " What did 
the first one do ? He wanted to please his daddy. He saw 
that the ink-pot was empty and thought he would Jill it. He 
made a big spot on his suit. And the second one ? He 
wanted to play with his daddy's ink, and he made a little 
spot. Which is the naughtiest ? The one who played with 
the ink pot. He was playing with it. The other one wanted 
to be kind. Did the one who wanted to be kind make a 
big spot or a little one ? He made a big spot, the other boy 
made a little one. Does it not matter the first one having 
made a big spot ? All the same, the other wanted more to 
do something wrong. The one who made a little spot wanted 
to do something more wrong than the other. 1 ' 

GROS (9) : " The one who wanted to be helpful, even if the 
stain is bigger, mustn't be punished." 

Nuss (10). The naughtiest is "the one who made the 
little stain, because the other one wanted to help." 

IIL THE STORY OF THE HOLES. Sci (6) repeats the 
stories as follows : ft The first one wanted to give her mother 
a surprise. She pricked herself and made a big hole in her 
frock. The second one liked touching everything. She took 
the scissors and made a little hole in her dress. Which one 
is naughtiest ? The one who wanted to take the scissors. 
She made a little hole in her frock. She is the naughtiest. 
Which one would you punish most, the one who made a 
little hole, or the other one ? Not the one who made a big 
hole ; she wanted to give her mother a surprise. 1 ' 

CORK (9). The naughtiest is " the second. She oughtn't 
to have taken the scissors to play with. The first one didn't 
do it on purpose. You cant say that she was naughty." 

These answers show what fine shades even some of the 
youngest children we questioned could distinguish and how 
well able they were to take intentions into account. The 
hypothesis may therefore be advanced that evaluations 
based on material damage alone are the result of adult con- 
straint refracted through childish respect far rather than 
a spontaneous manifestation of the child mind, Generally 
speaking, adults deal very harshly with clumsiness. In so 
far as parents fail to grasp the situation and lose then- 
tempers in proportion to the amount of damage done, in 
so far will the child begin by adopting this way of looking 
at things and apply literally the rules thus imposed, 


even if they were only implicit. And in so far as the 
parents are just, and, above all, in so far as the growing child 
sets up Ms own feelings as against the adult's reactions, 
objective responsibility will diminish in importance. 

With regard to stealing we also found two groups of 
answers, and here again, while both objective and sub- 
jective responsibility are to be found at all ages between 
6 and 10, it is the latter that predominates as the child 

Here are examples of objective responsibility : 

Sci (6) who showed signs of a subjective conception of 
responsibility in regard to clumsiness, changes his attitude 
here. He repeats the stories as follows : " A boy was with 
his friend. He stole a roll and gave it to his friend. A little 
girl wanted a ribbon, and put it round her frock to look 
pretty. Is one of them naughtier than the other ? Yes. 
. . . No. They're just the same. Why did the first one 
steal the roll ? Because his friend liked it. Why did the 
little girl steal the ribbon ? Because she was longing for it. 
Which one would you punish most ? The boy who stole 
the roll and gave it to his brother instead of keeping it for 
himself Was it naughty to give it ? No. He was kind. 
He gave it to his brother. Must one of them be punished 
more than the other ? Yes. The little boy stole the roll to 
give to his brother. He must be punished more. Rolls cost 

SCHMA (6) repeats the stories as follows : " There was a 
boy. As his friend had had no dinner, he took a roll and 
put in his pocket and gave it to his friend. A little girl went 
into a shop. She saw a ribbon. She says, it would be nice 
to put on my dress, she says. She took it. Is one of these 
children naughtier than the other ? The boy is, because 
he took a roll. It's bigger. Ought they to be punished ? 
Yes. Four slaps for the first. And the girl ? Two slaps. 
Why did he take the roll ? Because his friend had had 
no dinner. And the other child ? To make herself 

GEO (6) : " Which of them is the naughtiest ? The one 
with the roll, because the roll is bigger than the ribbon/' 
And yet Geo is like the other children perfectly well aware 
of the motives involved. 


(6) : " The little girl had a friend who had a cage and a bird. 
She thought this was too unkind. So she took the cage and 
Id the bird out. And the other one ? A little girl stoh a 
sweet and ate it. Are they both equally naughty or Is one 
of them naughtier than the other ? The one who stole the 
cage is naughtiest. Why ? Because she stole the cage. 
And the other one ? She stole a sweet. Is that one more 
or less naughty than the first ? Less. The sweet is 
smaller than the cage. If you were the daddy, which one 
would you punish most ? The one who stole the cage. 
Why did she steal it ? Because the bird was unhappy. 
And why did the other one steal the sweet ? To eat it" 

These cases of objective responsibility are thus all three 
of 6-year-olds. We found none above 7 years in the case 
of this kind of story. Here are some definite cases of sub- 
jective responsibility found in connection with the same 
stories. They are nearly all children of 9 and 10. The 
types are therefore better dissociated with regard to age 
than in the case of the stories about clumsiness. 

tells the two stories correctly. "What do you think 
about it ? Well, the little boy oughtn't to have stolen. He 
oughtn't to have stolen it, but to have paid for it. And the 
other one, she oughtn't to have stolen the ribbon either. 
Which of them is the naughtiest ? The little girl took the 
ribbon for herself. The little boy took the roll too, but to give 
it to his friend who had had no dinner. If you were the 
school teacher, which one would you punish most ? The 
little girl." 

Nuss (10) : " Which one is the naughtiest ? The little 
girl is became she took it for herself'' 

" Which one is naughtiest ? The one who steals the sweet. 
The first one took the cage so as to set the little bird free." 

CORM (9), G. : " It was good of the little girl who wanted 
to set the little bird free. The other one oughtn't to have eaten 
the sweet" 

GROS (9) : " The one who stole the sweet, that was 
naughtier. Why ? Because the other let the little bird go 
free again." 


Thus these answers present us with two distinct moral 
attitudes one that judges actions according to their 
material consequences, and one that only takes intentions 
into account. These two attitudes may co-exist at the 
same age and even in the same child, but broadly speak- 
ing, they do not synchronize. Objective responsibility 
diminishes on the average as the child grows older, and 
subjective responsibility gains correlatively in importance. 
We have therefore two processes partially overlapping, 
but of which the second gradually succeeds in dominating 
the first. 

What explanation can we give of these facts ? The 
objective conception of responsibility arises, without any 
doubt, as a result of the constraint exercised by the adult. 
But the exact meaning of this constraint has still to be 
established, because in cases of theft and clumsiness it is 
exercised in a rather different form from what appears 
in cases of lying. For in some of the cases we have been 
examining it is quite certain that adults, or some adults, 
apply their own sanctions, whether " diffused " (blame) 
or " organized " (punishment), in conformity with the 
rules of objective responsibility. The average housewife 
(most of the children we examined came from very poor 
districts) will be more angry over fifteen cups than over 
one, and independently, up to a point, of the offender's 
intentions. Broadly speaking, then, one may say that it 
is not only the externality of the adult command in 
relation to the child's mind that produces the effects we 
are discussing, it is the example of the adult himself. In 
cases of lying, on the other hand, we shall find that it is 
almost entirely in spite of the adult's intention that 
objective responsibility imposes itself upon the child's 


Restricted though the question under discussion may 
appear, it has a very distinct interest. When the adult 
allows himself to evaluate acts of clumsiness and pilfer- 
ing in terms of their material result, there can be no 
doubt that in most people's eyes he is unjust. On the 


other hand, those parents who try to give their children a 
moral education based* on intention, achieve very early 
results as is shown by current observation and the few 
examples of subjective responsibility we were able to note 
at 6 or 7. How is it, then, that in most of the cases under 
9-10 years the child accepts so completely the criterion of 
objective responsibility and even outdoes the average 
adult on this point ? The child is much more of an 
objectivist, so to speak, than the least intelligent parent. 
Also, most parents draw a distinction which the children 
precisely neglect to make : they scold, that is, according 
to the extent of the material damage caused by the 
chimsy act, but they do not regard the act itself exactly 
as a moral fault. The child on the contrary seems, as we 
have noted before, not to differentiate the legal or, as it 
were, the purely police aspect from the moral aspect of 
the question. It is " naughtier " to make a big spot on 
your coat than a small one, and this in spite of the fact 
that the child knows perfectly well that the intentions 
involved may have been good. To commit certain acts 
is therefore, in a sense, wrong in itself, independently of 
the psychological context. With regard to stealing, which 
is unanimously held up to children as a grave moral 
offence, this phenomenon appears even more clearly. 
Nearly all the children under 9-10, while paying full 
tribute to the thief's intentions, consider the theft of the 
roll and the cage a more culpable act both from a police 
and from a moral point of view than that of the ribbon or 
the sweet. Now, we can understand anyone condemning a 
theft regardless of the object pursued, but it is rather 
curious to see little children adopting an exclusively 
material criterion when they are asked to compare two 
such dissimilar acts as are described in our stories. 

The problem involved in all this is the following. 
What is the origin of this initial predominance of judg- 
ments of objective responsibility, surpassing in scope and 
intensity what may have been done or said to the children 
by adults ? Only one answer seems to us to be possible. 


The rules imposed by the adult, whether verbally (not to 
steal, not to handle breakable objects carelessly 3 etc.) or 
materially (anger, punishments) constitute categorical 
obligations for the child* before his mind has properly 
assimilated them, and no matter whether he puts them 
into practice or not. They thus acquire the value of 
ritual necessities, and the forbidden things take on the 
significance of taboos. Moral realism would thus seem to 
be the fruit of constraint and of the primitive forms of 
unilateral respect. Is this an inevitable product or an 
accidental result ? This is the point we shall try to settle 
in connection with lying. 

But before going too far in our generalizations, let us 
remember that the child's answers are given in answer to 
stories that are told to him and do not arise out of really 
experienced facts. As in the case of method we may 
therefore ask ourselves whether these verbal evaluations 
do or do not correspond with the child's real thoughts. 
These evaluations certainly change as the child grows 
older, and they also seem to be the result of some syste- 
matic influence. But are they a mere derivative, a verbal 
and therefore ineffectual deduction from the words spoken 
by adults, or do they correspond with a genuine attitude, 
moulded by unilateral respect and conditioning the child's 
behaviour before they inspire his sayings ? 

As we noticed in certain cases, the child pays far more 
attention to intentions where his own memories are con- 
cerned than when he is being questioned about one or 
other of our little stories. Such a fact as this surely shows 
us that if the child's objectivist attitude (unmistakable 
enough in his theoretical thought) corresponds to any- 
thing in his concrete and active thought, there 'must 
have been a time-lag taking place between one of 
these manifestations and the other, for the theoretical 
attitude is certainly a late-comer as compared to the 
practical. But the problem goes deeper than this, and 
the question may be raised whether at any moment in the 
immediate experiences of his moral life, or at any rate in 


those connected with clumsiness and lying, the child has 
ever been dominated by the notion of objective responsi- 

Immediate observation the only judge in the matter 
is sufficiently explicit on this point. It is very easy to 
notice especially in very young children, under 6-y years 
of age how frequently the sense of guilt on the occasion 
of clumsiness is proportional to the extent of the material 
disaster instead of remaining subordinate to the intentions 
in question. I have often noticed in the case of my own 
children, who have never been blamed for involuntary 
clumsiness, how difficult it was to take away from them 
all sense of responsibility when they chanced to break an 
article or soil some linen. Which of us cannot recall the 
accusing character which such a minor accident would take 
on as soon as it had happened, rising, with all the sudden- 
ness of a shock and overwhelming us with a sense of guilt 
that was the more burning, the more unexpected and the 
more irreparable the disaster. To be sure, all sorts of 
factors come into play (the sense of " immanent " justice, 
affective associations with previous carelessness, fear of 
punishment, etc.). But how could the material damage 
be felt as a fault if the child were not applying in a literal 
and realistic manner a whole set of rules, implicit and 
explicit, for which he feels respect ? 

We can therefore put forward the hypothesis that 
judgments of objective responsibility occurring in the 
course of our interrogatory were based upon a residue left 
by experiences that had really been lived through. Al- 
though new material may since have enriched the child's 
moral consciousness and enabled him to discern the 
nature of subjective responsibility, these earlier experi- 
ences are sufficient, it would seem, to constitute a per- 
manent foundation of moral realism which reappears on 
each fresh occasion. Now, since thought in the child 
always lags behind action, it is quite natural that the 
solution of theoretical problems such as we made use of 
should be formed by means of the older and more habitual 


schemas rather than the more subtle and less robust 
schemas that are in process of formation. Thus an adult 
who may be in the midst of reviewing aH Ms values and 
experiencing feelings of which the novelty surprises Mm, 
will, if he is suddenly faced with the necessity of solving 
someone else's problems, very probably appeal to moral 
principles wMch he has discarded for himself. For 
example, he will, if he is not given time to reflect, judge 
his neighbour's actions with a severity wMch would be 
incomprehensible in view of Ms present deeper tendencies, 
but wMch effectively corresponds to Ms previous system 
of values. In the same way, our children may perfectly 
well take account of intentions in appraising their own 
conduct, and yet confine themselves to considerations of 
the material consequences of actions in the case of the 
characters involved in our stories, who are indifferent to 

How, then, does subjective responsibility appear and 
develop witMn the limited domain we are analysing at 
present ? There is no doubt that by adopting a certain 
technique with their children, parents can succeed in making 
them attach more importance to intentions than to rules 
conceived as a system of ritual interdictions. Only the 
question is, whether tMs technique does not involve perpetu- 
ally taking care not to impose on their children any duties 
properly so called, and placing mutual sympathy above 
everything else ? It is when the child is accustomed to act 
from the point of view of those around Mm, when he tries 
to please rather than to obey, that he will judge in terms 
of intentions. So that taking intentions into account 
presupposes cooperation and mutual respect. Only those 
who have children of their own know how difficult it is to 
put this into practice. Such is the prestige of parents in 
the eyes of the very young child, that even if they lay 
down notMng in the form of general duties, their wishes 
act as law and thus give rise automatically to moral 
realism (independently, of course, of the manner in which 
the child eventually carries out these desires). In order 


to remove all traces of moral realism, one must place 
oneself on the child's own level, and give Mm a feeling of 
equality by laying stress on one's own obligations and 
one's own deficiencies. In the sphere of clumsiness and of 
untidiness in general (putting away toys, personal cleanli- 
ness, etc.), in short in all the multifarious obligations that 
are so secondary for moral theory but so all-important in 
daily life (perhaps nine-tenths of the commands given to 
children relate to these material questions) it is quite easy 
to draw attention to one's own needs, one's own diffi- 
culties, even one's own blunders, and to point out then- 
consequences, thus creating an atmosphere of mutual help 
and understanding. In this way the child will find himself 
in the presence, not of a system of commands requiring 
ritualistic and external obedience, but of a system of 
social relations such that everyone does his best to obey 
the same obligations, and does so out of mutual respect. 
The passage from obedience to cooperation thus marks a 
progress analogous to that of which we saw the effects in 
the evolution of the game of marbles : only in the final 
stage does the morality of intention triumph over the 
morality of objective responsibility. 

When parents do not trouble about such considerations 
as these, when they issue contradictory commands and are 
inconsistent in the punishments they inflict, then, obvi- 
ously, it is not because of moral constraint but in spite of 
and as a reaction against it that the concern with intentions 
develops in the child. Here is a child, who, in his desire 
to please, happens to break something and is snubbed for 
his pains, or who in general sees his actions judged other- 
wise than he judges them himself. It is obvious that 
after more or less brief periods of submission, during which 
he accepts every verdict, even those that are wrong, he 
will begin to feel the injustice of it all. Such situations 
can lead to revolt. But if, on the contrary, the child finds 
in his brothers and sisters or in his playmates a form of 
society which develops his desire for cooperation and 
mutual sympathy, then a new type of morality will be 


created in him, a morality of reciprocity and not of 
obedience. This is the true morality of intention and of 
subjective responsibility. 

In short, whether parents succeed in embodying it in 
family life or whether it takes root in spite of and in 
opposition to them, it is always cooperation that gives 
intention precedence over literalism, just as it was uni- 
lateral respect that inevitably provoked moral realism. 
Actually, of course, there are innumerable intermediate 
stages between these two attitudes of obedience and 
collaboration, but it is useful for the purposes of analysis 
to emphasize the real opposition that exists between them. 

ing with the judgments children make about lies we are 
penetrating a step further into the secret of childish 
evaluations. For there can be no doubt that lies present 
to the child's mind a far graver and more pressing 
problem than do clumsiness or even such exceptional 
actions as stealing. This is due to the fact that the 
tendency to tell lies is a natural tendency, so spontaneous 
and universal that we can take it as an essential part 
of the child's egocentric thought. In the child, therefore, 
the problem of lies is the clash of the egocentric attitude 
with the moral constraint of the adult. And this circum- 
stance will enable us to pursue on this point the analysis 
we outlined in connection with the early stages of the 
practice and the consciousness of rules in a game. 

The nature and diversity of childish lies are well known, 
thanks chiefly to the fine work done by W. Stern x and to 
the many other books which it inspired. Nor do we 
intend to pursue this line of enquiry. Our aim will be 
to undertake the analysis of the child's consciousness of 
lies, or, more precisely, of the manner in which the child 
judges and evaluates lying. We know why children lie. 
We know the educational difficulties which the question 
raises. The case is therefore exceptionally favourable to 

1 W. and C. Stem, Erinnerung, Aussage und Luge, 3rd. ed., 1922. 


the study of moral judgment In the child. If we can see 
in the course of an interrogatory how children think 
about lies and how they evaluate them, it will be relatively 
easy for us to decide whether the answers they give 
correspond with what we know from other sources of the 
practice of lying and truthfulness. 

From the first, childish evaluations appeared to us to 
be dominated by the notions of intentionality and ob- 
jective responsibility, and it is around this problem that 
we have grouped our examinations. Our interrogatory 
bore mainly upon the three following points : definition 
of a He, responsibility as a function of the lie's content, 
responsibility as a function of its material consequences. 
In addition to this we made a study of two points which 
will be dealt with in Section 4 : may children lie to one 
another ? and, why should one not lie ? 

The first question (definition of a lie) connects quite 
naturally with the problem of objective responsibility and 
moral realism. For what we have to find out is whether 
the child has understood that to tell a lie is wittingly and 
intentionally to betray the truth. Now even this prelim- 
inary question provoked the most suggestive answers and 
revealed the strength of realistic tendencies in the child. 

The most primitive and, at the same time, from our 
present point of view, the most characteristic definition 
we were able to find was a purely realistic one : a He is 
" a naughty word ". Thus the child, while perfectly well 
acquainted with a lie when he meets one, identifies it 
completely with the oaths or indecent expressions which 
one is forbidden to use. Here are examples : 

WEB (6) : " What is a lie ? It's when you say naughty 
things you oughtn't to say. What does ' naughty things ' 
mean ? Saying naughty words. Tell me a naughty word. 
Do you know any ? Charogne (Engl. lit., carrion). [This 
word is often used in Switzerland as an oath or a term of 
abuse and many people do not know its real meaning.] 
Is that a lie ? Yes. Why is it a lie ? Because it is a 
naughty word. When I say ' Fool ! ' is that a \tej- Yes. 
A boy once broke a cup, but he said he didn't do it. 


Was this a lie ? Yes. Why ? Because he had done it. 
You see this gentleman ? [a student], I say he is thirty- 
nine. Are you thirty-nine ? [The student answers, * No, 
only thirty-six.'] [To Web] Is it a lie to say he is thirty- 
nine when he is only thirty-six ? Yes, ifs a lie. Why ? 
Because he's thirty-six. Is it a naughty lie, or not ? 
Not naughty. Why ? Because it is not a naughty word. 
Is saying that 2 + 2 = 5 a lie, or not ? Yes. Naughty, 
or not naughty ? Not naughty. Why ? Because it is not 
a naughty word." 

LUD (6) : " Do you know what a lie is ? Ifs when you 
say naughty words. Tell me a naughty word that is a 
lie. . . . [Lud hesitates] Fire away. M . . . (Cam- 
bronne's word). Is it a lie ? Yes. Why ? Because it is 
a naughty word. I'm going to tell you a story. A boy 
once broke a cup, and then he said he hadn't done it. 
Was that a lie ? Yes. Why ? Because he said he hadn't 
done it. Is it a naughty word ? Yes. Why ? Because 
it was he had broken the cup'" 

Nus (6) : " What is a lie ? Ifs when you say naughty 
words. Do you know any naughty words ? Yes. Tell 
me one. Charogne. Is it a lie ? Yes. Why ? Because 
you mustn't say naughty words. When I say ' Fool ! ' is 
it a lie ? Yes" Nus also allows that the story of the 
broken cup involves a lie. 

RAD (6) : " A lie is words you mustn't say, naughty 

TUL (6). Same answers. " If I say 'fool' to someone, 
is it a lie ? Yes. Why ? Because it is a naughty word. 
And if he really is a fool is it a lie, or not ? Yes. Why ? 
Because it is a naughty word." 

Here finally is a child who hesitates between this 
definition and the right one. 

RIB (7) : " Do you know what a lie is ? Ifs when you 
H e ^ What is lying ? Ifs saying naughty words, When 
do people tell lies ? When you say something that isn't 
true. Is a lie the same thing as a naughty word ? No, 
ifs not the same thing. Why not ? They're not like each 
other- -Why did you tell me that a lie is a naughty word ? 
/ thought it was the same thing." 

This definition of a lie, of which we found many 
examples among the younger children, though it is by no 


means general and does not characterize a definite stage, 
is of great interest in fixing the children's initial attitude to 
lies. For nothing, after all, is more completely external 
to the moral consciousness, nothing is more like an tin- 
motivated taboo than an interdiction with regard to 
language. Why is one word perfectly proper, while 
another arouses everyone's indignation ? The child doesn't 
know in the least. He submits to the linguistic constraint 
and accepts this mystery without question. But this 
surely is the very type of those obligations which remain 
foreign to his practical understanding. How then does he 
come to identify lies with " naughty words " ? 

It should be noted in the first place that no mere verbal 
confusion is here at work. The child who defines a lie as 
being a " naughty word " knows perfectly well that lying 
consists in not speaking the truth. He is not, therefore, 
mistaking one thing for another, he is simply identifying 
them one with another by what seems to us a quaint 
extension of the word " lie ". Moreover, the relative 
frequency of this definition and the fact that this part of 
the interrogatory always came before the others would 
seem to indicate that this is no case of verbal suggestion 
due to the child's environment or to the interrogatory 
itself. There seems therefore to be only one explanation : 
to tell a lie is to commit a moral fault by means of 
language. And using naughty words also constitutes a 
fault committed by means of language. So that for the 
little child, who really feels no inner obstacle to the 
practice of lying, and who at six years old still lies more 
or less as he romances or as he plays, the two types of 
conduct are on the same plane. When he pronounces 
certain sentences that do not conform with the truth (and 
which his parents regard as genuine lies) he is astonished 
to find that they provoke the indignation of those around 
him and that he is reproached with them as with a fault. 
When he brings in certain expressive words from the 
street, the same thing happens. He concludes that there 
are things one may say and things one may not say, and 


he calls the latter "lies" whether they are Indecent 
words or statements that do not conform with fact. 

TMs identification of oaths and lies seems therefore to 
show how completely external to the child's own mind the 
interdiction of lying still remains in the early stages. 
This hypothesis agrees, moreover, with all that we know 
about the mind of the child under 7-8. Thus it will help 
us to see how perfectly natural are the <l objectivistic " or 
realistic evaluations which we shall find the younger 
children giving later about the stories they are asked to 

A more advanced definition of lies which remains the 
usual one until fairly late (between 6 and 10 on the 
average) consists simply in saying, " A lie is something that 
isn't true." But the mere words here must not deceive 
us, and we must get at the implicit notions which they 
conceal. We have seen elsewhere (J. R.) how difficult it is 
for the child to give an adequate definition of the notions 
he uses owing to his inability to realize them consciously. 
In order, therefore, to know the meaning of the definition 
we have just given we must ascertain whether the child 
confuses lying with every kind of inaccuracy (especi- 
ally with mistakes), or whether he implicitly considers a 
lie to have been told only when someone intentionally 
betrays the truth. In order to settle this point it will be 
sufficient to present to the child a certain number of 
stories, asking him each time whether a lie has been told or 
not (and why), whether there has been a mistake or not 
(and why), and whether there may not have been simul- 
taneously a lie and a mistake. 

The interrogatory shows that children of 5 to 7, while 
perfectly aware of the shade of difference between an 
intentional act arid an involuntary mistake, do not tend 
to stress this distinction at all, and often, on the contrary, 
group both facts under the same name of " a lie ". 

Here are examples of this type of definition, which 
therefore constitutes the second type, if we regard as a 
first type that which identifies lies and naughty words : 


CLAI (6) : " Do you know what a lie is ? If s when you 
say what isn't true. Is ' 2 -f 2 - 5 ' a lie ? Yes, it's a tie. 
Why ? Because it isn't right. Did the boy who said 
that ' 2 + 2 5 * know it wasn't right or did he make a 
mistake ?He made a mistake. Then if he made a mistake, 
did he tell a lie or not ? Yes, he told a lie. A naughty one ? 
Not very. You see this gentleman [a student] ? Yes. 
How old do you think he is ? Thirty. I would say 28. 
[The student then says he is 36.] Have we both told a 
lie ? Yes, both lies. Naughty ones ? Not so very naughty. 
Which is the naughtiest, yours or mine, or are they both 
the same ? Yours is the naughtiest, because the difference is 
biggest. [Cf. moral realism.] Is it a Me, or did we just 
make a mistake ? We made a mistake. Is it a lie all the 
same, or not ? Yes, it's a lie" 

WEB (6) already examined in connection with " naughty 
words " answers towards the end of the interrogatory like 
the last child : " Is making mistakes the same thing as 
telling lies ? Not the same thing I'm going to tell you a 
story. Once there was a boy who didn't know where the 
Rue des Acacias was (the street where Web lives). A 
gentleman asked him where it was. The boy answered 
' I think it's over there, but I'm not sure/ And it wasn't 
over there ! Did he make a mistake, or did he tell a lie ? 

ft was a lie. Did he make a mistake or not make a 

mistake ? He made a mistake. Then it wasn't a lie ? 
He made a mistake and it was a lie. 3 ' This last answer 
should be especially noted for it shows how for some 
children the two notions overlap. 

MAS (6) : " What is a lie ? When you talk nonsense. 
You tell me about something that is a lie. A boy^ once 
said he was a little angel, and it wasn't true. Why did he 
say that ? For a joke. Are we allowed to tell lies ? 
j\fo Why not ? Becaiise it's a sin and God doesn't want 
us to sin. A boy told me that 2 + 2=5. Is it true ? 
No, it makes 4. Was it a lie or did he make a mistake ? 
He made a mistake. Is making a mistake the same as 
telling a lie or different ? The same. Look at me. I'm 
30. A boy told me I was 60. Was it a lie or did he make 
a mistake ? It was a lie. Why ? Because what he said 
was a sin." " Which is naughtiest, to make a mistake or 
to tell a lie ? Both the same.'' 

CHAP (7) : " What is a lie ? What isn't true, what they 
say that they haven't done. Guess how old I am. Twenty. 
No, I'm thirty. Was it a lie what you told me ? / 


didn't do it on purpose. I know. But is It a He all the 
same, or not ? Yes, it is all the same, because I didn't say 
how old you really it-ere. Is it a He ? Yes, because I didn't 
speak the truth. Ought you to be punished ? No. Was 
It naughty or not naughty ? Not so very naughty, Why ? 
Because I spoke the truth afterwards I ** 

These observations yield two conclusions. The first is 
that in practice the children can more or less distinguish 
between an intentional act and an involuntary error. The 
idea of intention (not just the intentional act but the idea 
of the intentional act, which is not the same thing) appears 
roughly at the same time as the first " whys ", that is, at 
about the age of three. Nevertheless, during the next few 
years the child does not distinguish between intentional 
actions and others as clearly as we do. The proof of this 
is that his thinking is finalistic, animistic and artificialistic 
to a degree that the average adult does not suspect, and 
this is so precisely because he fails to dissociate involun- 
tary, unconscious and mechanical movements from con- 
scious psychological action (see C. IF.) . There was therefore 
some reason to doubt whether a child of 6-7 could really 
distinguish an involuntary error from an intentional lie. 
The answers we have just recorded seem to show that the 
distinction is, at the best, in process of formation. 

Only and this is our second conclusion these two 
realities are still not dissociated on the plane of moral 
reflection. Mistakes, although they are distinguished from 
lies proper, are still conceived as constituting lies. More 
precisely, a lie is defined in a purely objective manner, as 
an affirmation that does not conform with fact, and even 
if the child can recognize two types of statement those 
which are, and those which are not intentionally false, he 
subsumes them both kinds under the category of " lies ". 
TMs identification is analogous to that examined above 
between lies and " naughty words ". But in this par- 
ticular case the identification is probably helped by the 
child's lingering inability to dissociate between the ideas 
of an intentional act and of an involuntary act. For it 


may very well be that these ideas remain undififerentiated 
on the plane of reflection longer than they do on the 
plane of practice. This would supply us with one more 
example of that time-lag between action and thought of 
which we have already spoken. In support of this view 
it should be noted that the identification of mistakes and 
lies disappears towards the age of 8, i.e. at the same time 
as the bulk of the phenomena of animism, artificialism, 
etc., i.e. at the same time as the other signs of inability to 
dissociate the idea of the intentional from the idea of the 

Whether these interpretations are right or not, it is 
obvious that both the answers we have been discussing and 
the first type of definitions reveal in the child a tendency 
to consider lies in a purely realistic manner and inde- 
pendently of the intentions involved. This will be amply 
confirmed by the remainder of our analysis. 

Let us now turn to the third type of definitions, or 
correct definitions of lies : any statement that is in- 
tentionally false is a lie. Not till about the age of 10-11 
do we find this definition in an explicit form. But among 
the children who begin by saying "A lie is something 
that isn't true " the majority are actually thinking of 
deceit, that is to say of the intentional deed. Here are 
examples, beginning with such cases of implicitly correct 

ZER (6), G. : " What is a lie ? . . .Do you know 
this word ? Yes. What does a lie mean ? . . . A boy 
knocked over a chair and then he said he hadn't done it. 
Was it a lie ? Yes. Why ? Because he had done it. I 
say you are called Helen, is it true ? No, I'm called 
Madeleine. Is it a lie ? No. Why ? . . . A little boy 
told me that 2 + 2=5, Is it a lie, or did he make a 
mistake. He made a mistake. Is it a lie all the same ? 
No, he made a mistake." 

GAR (6) : " What is a lie ? It's disobeying. If you 
don't go into the garden when you've been asked to go, is 
it a lie ? No, M'sieu. Then what is a lie ? When you do 
something wrong and then you don't tell. If a boy breaks a 


cup and then says it was the cat, is it a lie ? Yes, M'sieu. 
A boy told me that 24-2 = 5. Is that a lie ? It's not a 
lie. Then what did he do ? He wasn't paying attention" 

EEC (7) : " What is a lie ? When you say something 
and then afterwards it isn't true. A child doesn't know the 
names of the streets (in Geneva). A gentleman asks Mm 
where the Rue de Carouge is, and the child tells Mm 
wrong. Is that a lie, or has he made a mistake ? He 
didn't tell a lie. Why not ? Became he -made a mistake 
[Cf. the * because ']. Why did he make a mistake ? 
Because when you make a mistake it's that you don't know, 
but a lie you know, but you say it isn't true [ = you don't 
say what you know]/' 

KE (7) : " What is a lie ? When you don't say things 
right. Is 24-2=5 a lie ? No, that's not one. Why did 
the boy who said it was tell me so ? Perhaps he made a 
mistake. What is making a mistake ? When you don't 
say things right. Is making a mistake the same tMng as 
telling a lie ? No. What is the difference ? Making a 
mistake is when you don't say things right ; a lie is true, 
only you don't say it [ = you know the truth but do not 
say it]. And a joke (Fr. une blague) ? It's when you say 
something for a lark." 

PUT (8). A lie is " when you say what is not true. But 
when Pui guesses our age wrong he says it is not a lie. 
Why ? Because I didn't know. 33 

Here are some good explicit explanations. 

LAU (8) : " A boy who tells a lie knows what he is doing, 
but he doesn't want to say it. The other one [who makes a 
mistake] does not know" 

ARL (10) : " When you lie you're doing it on purpose, 
when you make a mistake, you don't know!' 

KEI (10) : " A lie is when you deceive someone else. To 
make a mistake is when you make a mistake [deceive 
yourself] (Fr. quand on se trompe)." 

This preliminary analysis of the definitions brings out 
the initial difficulty experienced by the child under 8 in 
understanding the true nature of a lie. Naturally inclined 
as he is to think about himself rather than about others, 
the child does not see the full significance of deceit. He 
lies as he romances, so that the obligation not to lie, 
imposed upon him by adult constraint, appears from th 


first in its most external form : a He is what does not 
agree with the truth independently of the subject's in- 

Let us now turn to the second and third parts of our 
interrogatory, that is, to the evaluation of the stories in 
terms of the content and of the consequences of the lies. 
In order to solve these questions we made use as before of 
stories to be compared, and before examining the answers 
we obtained, we must begin by discussing the significance 
of these stories. 

The question of the evaluation of lies from the point of 
view of their content is the problem of responsibility. 
Responsibility is here evaluated either as a function of 
the aim of the lie (subjective) or as a function of its 
degree of falseness (objective). The difficulty is therefore 
to find stories in which the lie leads to no material result 
and is not accompanied by any material circumstance 
whatsoever. For experience showed us that any lie told 
in connection with an act of clumsiness or of something 
analogous (breaking a cup and saying it was the cat, 
losing the scissors and denying having done so, etc.) is 
invariably considered worse than a lie that is not con- 
comitant with action of this sort. The child cannot, in 
such cases, dissociate the lie from the accompanying act 
and therefore judges it in terms of the consequences 
of the act. These features will reappear later and will 
confirm what we hinted at in the last section. But for 
the moment they do not concern us. It is the lie itself in 
its actual content that we want the child to evaluate. 

After much groping and hesitation we selected the three 
following pairs of stories. Each pair contains a lie or 
mere inaccuracy devoid of any evil intention but marking 
a glaring departure from fact, and a lie of quite probable 
content, but told with the manifest intention to deceive. 

I. A. 1 " A little boy [or a little girl] goes for a walk in 
the street and meets a big dog who frightens him very 

1 Mile Descoeudres (loc. cit.) is responsible for this story. 


much. So then he goes home and tells Ms mother he has 
seen a dog that was as big as a cow." 

B. " A child comes home from school and tells Ms 
mother that the teacher had given Mm good marks, but it 
was not true ; the teacher had given him no marks at all, 
either good or bad. 1 Then his mother was very pleased 
and rewarded him." 

II. A. "A boy was playing in Ms room. His mother 
came and asked him to run a message for her. But he 
didn't feel like going out so he told his mother Ms feet 
were hurting. But it wasn't true ; his feet were not 
hurting him in the least." 

B. "A boy wanted very much to go for a ride in a 
car, but no one ever asked him. One day he saw a lovely 
motor car in the street and would have loved to be inside 
it. So when he got home he told them that the gentleman 
in the car had stopped and had taken him for a little 
drive. But it was not true ; he had made it all up." 

III. A. "A boy couldn't draw very well but he would 
have liked very much to be able to draw. One day he 
was looking at a lovely drawing that another boy had 
done, and said : ' I did that drawing V 

B. " A boy was playing with the scissors one day 
when Ms mother was out and he lost them. When Ms 
mother came in he said that he hadn't seen them and 
hadn't touched them/' 

Here, finally, are the stories we used for the analysis of 
the children's evaluation in terms of the material results 
of the lies. 

IV. A. " A child who didn't know the names of streets 
very well was not quite sure where the Rue de Carouge 
was [a street near the school where we were working]. 
One day a gentleman stopped him in the street and asked 
him where the Rue de Carouge was. So the boy an- 
swered, ' I think it is there.' But it was not there. The 
gentleman completely lost Ms way and could not find the 
house he was looking for." 

B. " A boy knows the names of the streets quite well. 
One day a gentleman asks Mm where the Rue de Carouge 
is. But the boy wanted to play him a trick and said, it 

1 The child must not be said to have had a bad mark, otherwise 
the subject will regard him as ipso facto " naughtier " independently 
of the superimposed lie. 


is there, and showed him the wrong street. But the 
gentleman didn't get lost, and managed to find his way 

V. We also made use of several stories in which the lie 
is accompanied by clumsiness (breakages, stains, etc.). 

We tell the children the stories, two at a time. Before 
questioning the subject it is best to make him tell the 
stories from memory so as to make sure that he has 
understood them and especiaEy that he has grasped 
the intentions that come into play. This is done by 
asking the child after each story : " Why did the boy say 
that ? " In general our subjects understood at once that 
story LA involved a simple exaggeration due to fear, 
stories II. B and III. A mere inventions due to desire, 
etc., whereas stories I. B and III. B related to genuine 
acts of deceit due to the desire for sweets, to laziness or 
the fear of a deserved punishment. Then having made 
sure that the child has understood, you ask him to com- 
pare the two stories and to say which of the two lies or of 
the two boys in question was " the naughtiest " and why. 

The results obtained by means of this technique were 
definite and suggestive. In the matter of clumsiness and 
theft the child, it will be remembered, seemed to judge 
of actions from their most external aspect (their results) 
before taking any account of the intentions in play. The 
analysis of the evaluations relating to lies not only con- 
firms this conclusion but allows us to go beyond it and 
to say that even apart from all consideration of the 
material consequences of actions the child's mind remains 
set in the direction of objective responsibility. 

For our stories I to III reduce the part played by 
material circumstances to the minimum ; so that when 
the child answers he will not be thinking of the actions 
about which the lie was told, but of the lie itself and 
of the liar's intentions. Now, what struck us was the 
fact that a large number of the younger children whom 
we questioned evaluated the lies, not according to the 
intentions of the liar, but according to the greater or 


lesser likelihood of the lying statement. In point of fact, 
judging by the first two types of definition which we have 
just been studying, we had every reason to expect that the 
more a statement departed from truth, the more it would 
seem to the child to be a lie. And this in fact is what we 
now see to be the case. But the extraordinary thing is 
that the judgment of value itself, i.e. the fact of evaluating 
the lies or liars as ll naughty " or " not naughty " should 
obey the same principle. 

And yet this is what happens on the verbal and abstract 
plane on which our interrogatories are unfortunately 

Here are examples concerning our first three pairs of 



the two stories correctly : " Which of these two children 
is naughtiest ? The little girl who said she saw a dog as 
big as a cow. Why is she the naughtiest ? Because it 
could never happen. Did her mother believe her ? No 
because they never are [dogs as big as cows], Why did she 
say that ? To exaggerate. And why did the other one 
tell a lie ? Because she wanted to make people believe that 
she had a good report. Did her mother believe her ? 
Yes. Which would you punish most if you were the 
mother ? The one with the dog because she told the worst 
lies and was the naughtiest.' 1 

BUG (6) : " Which is naughtiest ? The one with the 
cow. Why is he the naughtiest ? Because it isn't true. 
And the one of the good marks ? He is less naughty. 
Why ? Because his mother would have believed, because she 
believed the lie. [This is not a slip. We have met with 
many cases of children of 6-7 who, like Fel and Bug, 
measure the naughtiness of a lie by the degree of its 
incredibility to adults. Consequently the lie about the 
good marks is not so bad because the mother will be 
easily taken in by it !] And why did the other child tell 
the lie about the cow ? Because he was telling his mother a 
lie. Which would you punish most ? The one who said 
he saw a dog as big as a cow" 

MAE (6) : " Which of the two is the worst lie ? The 
dog who was as big as a cow. And if you had to punish 


those children, which would you punish most ? The one 
who said he saw a dog as big as a cow. Why did the other 
one say that the teacher had given him good marks ? 
Because he had been naughty and didn't dare say so." 

JLE (7) : " Which was the worst lie ? The one who said 
he saw a dog as big as a cow. Why is it the worst ? 
Because there's no such thing (Fr. Parce que fa riexiste pas). 
[Note this formula which expresses very well the point of 
view of objective responsibility in the case of lies.] 
Which of the two children should be punished most ? 
The one who saw a dog as big as a cow. Why ? Because 
it isn't true. It's not nice. Why is it not nice ? Because 
it isn't true. Why did the other one say that his teacher 
had given him good marks ? So that his mother should 
buy him something. Why did the other one say he had 
seen a dog as big as a cow ? Just like that. Because he 
wanted to say something like that [to boast of having seen 
something wonderful]. Why did he say something like 
that ? Because he saw a cow, perhaps. He didn't see 
properly.'' Ke therefore understands the two children's 
respective intentions perfectly well ; but he sticks none 
the less to his objective evaluation of their responsibility. 

Roc (7) : " What do you think of these two lies ? 
Are they just the same, or is one worse than the other ? 
One is worse than the other. Which ? The one who saw a 
dog as big as a cow. Why ? No one ever saw dogs as big 
as cows. If you were the mother, which would you 
punish most ? The one who told the lie about the cow." 

BURD (7) : " The naughtiest " is the one who saw a dog 
as big as a cow. It is naughtier because his mother knew 
[that it was false or impossible], whereas the other one, the 
mother didn't know. If you say something that mother 
doesn't know, it is less naughty because his mother might 
believe. If the mother knows it isn't true, then it's a bigger 
lie." It would be impossible to express more clearly the 
idea that the moral gravity of a lie is to be measured 
solely by the improbability of the lying statement : what 
mother believes is not naughty whereas the immediately 
obvious lie is naughty ! 

DRIV (7) : " Which of these two boys is the naughtiest ? 
The one who said he saw a dog as big as a cow. Is that 
naughtier ? He is the naughtiest because he told the biggest 
lie. Why is it the biggest lie ? Because he said a much 
bigger one than in the other story. Why? Because it 
couldn't have happened." 


SAY (7) says that the naughtiest Is the lie about the 
dog " because it couldn't be. And the one about the 
teacher ? Yes, that could be, but it's not true. Which of 
these two children would you punish most ? The one of 
the dog. Why ? Because it couldn't happen. Did the 
mother believe the lie about the dog ? JVo. And would 
the mother have believed the lie about the teacher ? 
Perhaps. When is it naughtiest, when the mother be- 
lieves the lie, or when she can't believe it ? When you 
can't believe it.* 3 

TROT (7) : " Which one should be punished most ? 
The one who saw a dog as big as a cow. Why is he the 
naughtiest ? Because a dog is never as big as a cow. Did 
the mother believe it ? No, she didn't believe it. And did 
the mother believe the lie about the teacher ? Yes, 
sometimes you are good, sometimes you are not. WMch of 
the two children would you punish most ? The one with 
the dog. Why ? Because it doesn't exist [a dog as big as 
a cow] and because at school sometimes you get punished, 
sometimes you don't. 9 ' [The second lie is therefore more 
plausible and consequently less naughty.] 

BOH (9) : " Which is the naughtiest ? Both the same. 
If you were the father, which would you punish most ? 
The one who said he saw a dog as big as a cow. Why ? 
Because it couldn't be." 


" WMch is the naughtiest ? The one of the car. Why ? 
IPs a bigger lie than the other. Why ? He tells more of 
a lie than the other." Note this quantitative expression 
[meaning that what he says is falser than] which brings 
out the point of view of objective responsibility. 

Roc (7) : " Which is naughtiest ? The one of the car. 
? Because you can't take little girls for drives in cars." 

PIE (7). The story of the sore feet " is a lie. Why ? 
Because he wanted to stay in. And the story of the motor 
car ? That's a lie too. Why ? Because he wanted to say 
that. Why did he want to say that ? Because he wanted 
to take them in (Fr. faire le malin). Which was the 
naughtiest ? The one of the car. Why ? Because it was 
a bigger lie. Why ? Because it was a longer sentence. It 
is bigger. It's naughtier. Why ? Because he told his 
mother lots of things. Not just one sentence, but a long one 
[ - a complicated story], If you had to punish them 


which would you punish most ? The one of the car." 
This shows Pie's purely quantitative conception of the 
* matter : the longer the story, the worse the lie ! 

BURD (7 ; 10) : " Which is the naughtiest ? The one 
with the car. Why ? Because the mother knew he couldn't 
have gone in such a short time [ go for a drive during his 
short absence]." 

DRIV (7 ; 6) : " Which of the two boys was the 
naughtiest ? The one who said he went in the car. 
Why ? It was the biggest lie. Why do you think it was 
the biggest lie ? Because he said he "had been in a car and 
it was not true. And the other one who said his feet hurt, 
was that just as naughty or less naughty ? Less naughty. 
Why ? Because he said his feet hurt him. And was it 
true ? No, M'sieu. Then why do you think it was a less 
naughty lie ? ... Why did he go and say his feet hurt 
him ? So as not to go [the messages]. And why did the 
other one go and say he had been for a drive ? To make 
his mother believe he had. Which is the naughtiest ? 
The one who said he had been in a car" 

DEPR (8) : " Which was the naughtiest ? The one who 
said he done the drawing. Why is he the naughtiest ? 
Because everyone thought it wasn't him, and he said it was 
him. Why did he say that ? So that everyone should 
think it was him. And the boy with the scissors, why did 
he tell a lie ? So that his mother should not punish him. 
If you had to punish these children, which would you 
punish most ? The one who said he had done the drawings. 
Why would you punish him most ? Because he told a 
worse lie" 

GREM (8 ; 10) : " Which is the naughtiest ? The one 
who said he had done the drawing. Why ? Because he 
couldn't draw, and he said it was him. Why did he tell 
this lie ? To boast. And why did the other one lie ? 
He thought he would rather play than go and fetch the 
scissors. Which one would you punish most ? The one 
who said he had done the drawing" 

PIT (9) : " Which is the naughtiest ? The one with the 
drawing because he didn't know how to draw and he said he 
had done it. Why did he say that ? Because he wanted 
to say it was him, just as though the others didn't know that 
he couldn't draw. And why did the other one tell a lie ? 
He wanted to put the blame on someone else. Then which 
one would you punish most ? The one with the drawing." 


The general principle underlying these answers is clear. 
The more unlikely the lie, the more its contents mark a 
departure from reality, the worse it is. The lie about the 
dog that was as big as a cow is particularly naughty 
(Main x ) " Because it never could be *\ li Because there's 
no such thing", because " No one ever saw dogs as big 
as cows", because "It's a bigger lie". And, above all, 
the essential point is that it is a lie because te you can't 
believe it ". Mother will see straight away that it is false 
and the lie will be exposed in the presence of all. But 
there is nothing unusual about having good marks at 
school. It is quite likely to happen, and parents will 
readily believe it. It is therefore only a little lie, all the 
more innocent because a mother is taken in by it. In the 
same way, the lie about the motor car is bad because 
" you can't take little girls for drives in cars " and 
" because it is a longer sentence ", therefore a story of 
some importance ; whereas to complain falsely of sore feet 
so as not to have to run messages is " almost nothing for 
the mother " in so far as she is taken in by it. Finally, 
to boast of being able to draw a picture when you can't 
draw at all is all the worse because of the fact that every 
one is able to verify the unlikelihood of the statement. 

We are therefore in the presence of judgments of ob- 
jective responsibility in an unadulterated state, or at any 
rate in a much simpler form than those whose contents 
we analysed in connection with clumsiness and stealing. 
In those cases it was always a question whether the child 
was not fascinated by the material aspect of the action, 
by the purely physical damage done to the adults ; 
whereas in the present cases the material element is 
reduced to its minimum. We have on the one hand lies 
whose intention is manifestly interested, lies that really 
" deceive ", as the child himself admits. On the other 
hand we have simply romancing, jokes (" blagues ") or 
exaggeration, as the child himself realizes. And yet the 

1 The term vilain is that used more particularly by Genevan children 
to mean " wicked " or " immoral ". 


subjects we have quoted ignore the liars' intention, and, 
basing themselves only on the degree of likelihood of the 
lie, judge it from the most external and objectivistic 

There can therefore be no question of explaining these 
facts by any inability on the children's part to understand 
the stories. As can be quite clearly recognized, all the 
subjects whose answers we have quoted understood the 
intentions that entered into the matter. It is not from 
lack of psychological penetration that they evaluated lies 
according to the criterion of objective responsibility ; it is 
because intention does not seem to them to count from 
the point of view of morality itself. 

But we must hasten to add that even on the plane of 
verbal reflection, where our interrogatories take place, the 
notion of objective responsibility is not held in all its 
purity by any one child. It is always mixed with sub- 
jective responsibility. A given child, who pays no 
attention to the intentions involved in the story of the 
dog, will, on the contrary, judge the stories of the drawing 
and the scissors in accordance with the psychological 
context, and so on. There can therefore be no question 
of two real stages. All we can say is that objective 
responsibility is a phenomenon frequently to be found 
among the younger children but subsequently diminishing 
in importance. 

Psychological statistics are notoriously open to sus- 
picion, but the following little calculation may be of value 
in this connection as showing how objective responsi- 
bility changes with age. Taking as our unit not each 
separate child since one and the same child may give 
answers according to both types of responsibility taking 
as our unit, then, the answer given by each child in 
connection with each story, and dividing those units 
according to the two types of responsibility, objective and 
subjective, we find that the average age for the objective 
type is 7 while for the subjective it is 10. As these 
statistics deal only with children from 6 to 12 it will be 


seen that, broadly speaking, objective responsibility dis- 
appears in favour of subjective responsibility as the child 
advances in age. 

And now, in order to get at the real significance of the 
preceding answers, let us by means of the same pairs of 
stories examine the reactions of children whose minds are 
directed towards the idea of subjective responsibility. 


" Which is the naughtiest ? Both ike same, Exactly the 
same ? One is a little naughtier. Which ? The one who 
said the teacher had said he -was very good. Why did he say 
that ? So that they should give him something. And the 
other one ? . . . . If yon were the father, which would 
you punish most ? The one of the teacher. Why ? 
Because he is the naughtiest." 

LOUR (8) : " Which is the naughtiest ? The one who 
told about the teacher. Why is he the naughtiest ? 
Because a dog like that doesn't exist and because it is 
naughtier to tell a lie. [Thus Lour reverses the argumenta- 
tion used by the previous subjects, and from the fact that 
a dog is never as big as a cow he concludes that to ex- 
aggerate in this way is not to lie.] Why did the one with 
the teacher tell a lie ? So that Ms mother should- reward 
him. And the other one ? For a joke (Fr. blague). 
Which would you punish most ? The one who said what 
the teacher hadn't said." 

CHRA (8) : The one who said he had seen a dog as big 
as a cow is less naughty " because it was as a joke. The 
one of the teach&r is naughtier because he had said it so that 
his mother should not scold him." 

DEP (8 ; 9) : " Which is the naughtiest ? The lie about 
the teacher. Why ? Because the story about the dog is 
nothing, but the story about the teacher made the mother 
angry. Why did the first boy say he had seen a dog as 
big as a cow ? Just for swank (Fr. Pour faire le petit 
crdneur). And the other one ? So that his mother should 
not punish him. Which of these two children would you 
punish most ? The one with the teacher. Why ? Because 
absolutely the opposite [of the truth]. The one about the 
dog the little boy told his mother and father for a lark (Fr. tine 
farce). Why did he tell it ? Perhaps because it was a little 
calf. 9 ' 

PIT (9 ; 3) : The naughtiest is the one of the teacher 


" because he said he had been very good so that his mother 
should give him some pennies or some chocolate. -And why 
did the other one say he had seen a dog as big as a cow ? 
Because he thought a cow was a dog and because his brain 
wasn't working properly.' 3 

DELE (9) : " The naughtiest is the one of the teacher. 
Why is he naughtier than the other ? Because the other 
one, it wasn't possible. Why did he say it was ? Because 
the dog was very big. And why did the other one say that 
the teacher had given him a good mark ? So that his 
mother should pel him. 31 

ARL (10) : The naughtiest is the one " who deceived his 
mother by saying that the teacher was pleased. Why is he 
the naughtiest ? Because the mother knows quite well that 
there aren't any dogs as big as cows. But she believed the 
child who said the teacher was pleased. Why did the child 
say the dog was as big as the cow ? To make them 
believe it: As a joke. And why did the other one say that 
the teacher was pleased* ? Because he had done his work 
badly. Was that a joke ? No, it is a lie. Is a lie the 
same thing as a joke ? A lie is worse because it is bigger." 

KEI (10) : " Which is the naughtier ? The lie about the 
teacher. Why ? Because he deceived his mother. But so 
did the other. But he [the one of the teacher] had said 
something the teacher hadn't said. The other one too had 
said something that wasn't true. [Our counter-sug- 
gestions, no matter how insistent, are ineffectual.] He 
had told a great big lie. The teacher hadn't said he was good. 
Why is it naughtier than the lie about the dog? 
Because you can see better that it [the lie about the dog] is 
not true. You can't tell with the lie about the teacher. Why 
did he say he had seen a dog as big as a cow ? To make 
them believe he had seen something marvellous." 

Ros (n) : The lie about the teacher " is a bigger lie 
because it was a bad thing to do. 1 ' 

SAV (7!) : " Which is the naughtiest ? The one with 
the sore feet, I think. Why did he say that ? So as 
not to go out. And the other one ? To take them in 
[as a joke]/' 

FER (8) : The naughtiest is the one who pretends his 
feet hurt " because he didn't want to go out and because he 
disobeyed his mother." 

LOURD (8) begins by saying that the naughtiest is the 
one of the car. Then he adds spontaneously : " I'm 


wrong. I would punish the one wouldnt ^o out. And 
what did the other one do ? That was for a joke " 

Sci (9 ; 7) : " Which is the naughtiest ? The one with 
the feet that hurt. Why ? Because it wasn't true. Why 
did he say that ? So as not to go out, And the other 
one ? Xo make them believe he had been in a car once. 
Which is the naughtiest ? The one with the bad feet." 

FER (8) : " Which is the naughtiest ? The first [drawing]. 
He said it to swank. And the second one [scissors] so as not 
to be scolded. Quite right. Which is the naughtiest ? 
The second." 

LOURD (8) : " Which is the naughtiest ? The one of the 
scissors, because he said that so as to keep them. And the 
other ? He said it so thai people should praise Mm. 
Which would you punish most ? The one of the scissors. 
Why not the other one ? Because the boys in Ms class 
know quite well how he draws." Lourd therefore thinks 
that the more obvious a lie is, the less it matters the 
converse of what was maintained by the votaries of 
objective responsibility. 

DELA (9) : The naughtiest is the one of the scissors. 
" Why did he say that ? So that his mother shouldn't 
scold him. And why did the other boy say that he had 
done the drawing ? To make the bdys think that he could 
draw well." 

What strikes one about these answers is that the very 
arguments used by the partisans of objective responsi- 
bility reappear but in support of the opposite thesis. The 
previous set of subjects regarded the lie about the dog 
and the cow as serious because it was unlikely, and the 
proof brought forward by the child was that the mother 
" didn't believe it " and " saw at once that it was a lie " 
The present subjects, on the contrary, regard the same 
circumstances as indicating that the lie is not serious : if 
you see straight away that a statement is false, it means 
that the person who makes it is not trying to deceive you 
but is exaggerating or making a mistake. Thus Arl does 
not consider that the story about the dog is a lie, since the 
mother knows that there are no dogs as big as cows. The 
less a lie appears to be one, the worse it is, which is the 
direct opposite of the view held by the younger children. 


Let us now turn to the last point which we have to 
examine, namely objective responsibility as a function of 
the material consequences of the lie. In using stories 
invented for the preceding experiments we found that the 
child more or less consistently regarded a lie as all the 
more serious if it accompanied actions that had re- 
grettable (material) results. Thus consequences outweigh 
intentions, as was the case in connection with clumsiness 
and stealing. For example : 

QUEL (7) : "A little boy upset an ink-pot on the table. 
When his father came back the boy told him it wasn't he 
who knocked over the ink-pot. He told him it was the 
cat ! Was it a lie ? Yes. Naughty or not naughty ? 
Naughty. Why ? Because it made a big stain.' 3 

DUB" (8) : "Two children bought some eggs for their 
mother. But they played on their way home and broke 
the eggs. The first child broke twelve eggs, the other 
child, one. When they got home they told their mother 
that a big dog had jumped at them and broken the eggs. 
Was that telling lies ? Yes. Were both the lies equally 
naughty ? No. One was worse than the other. Which ? 
The first, beca^tse he broke more. But I am not talking 
about the first child, I mean the first lie. What was the 
first lie ? He said it was the dog. And the second. He 
also said it was the dog. Then are both these lies equally 
naughty ? No, the first is because the boy broke most eggs" 

In other words, when a material action is too closely 
attached to the lie, the child has a certain difficulty in 
dissociating the lie in so far as it is psychological action 
from the actual results of the concomitant external act. 
The only interest of these examples is that they bring out 
this lack of dissociation. It would be very difficult to 
carry the analysis of children's evaluations any further by 
means of stories of this kind because the data in question 
are not two, but three in number : the lie, the concomitant 
act and the results of the act. We therefore abandoned 
this type of question. 

Instead, we tried to solve the problem systematically by 
presenting to the children a pair of stories, one of which 
contained a clearly intentional piece of deceit having no 


appreciable physical consequences, the other a simple 
mistake, but leading to harmful results. These are stories 
IVA and IVs given at the beginning of this section : i A 
boy purposely shows a gentleman the wrong way, but the 
gentleman does not get lost. 2 C A boy shows a gentleman 
the wrong way by mistake and the gentleman gets lost. 
The result of the interrogatory is that here, as in the case 
of clumsiness and stealing, some children think about the 
material consequences and evaluate the lie from this point 
of view only ; others think only of the intention. The 
average age of the first is 7, that of the second is 9. Thus 
we have here two distinct attitudes, one of which seems 
to be more evolved than the other, but examples of both 
are to be met with at any age between 6 and 12. 
Here are examples of objective responsibility. 

THE (6) : " Which of the two is the naughtiest ? The 
one who didn't know where the Rue de Carouge was. Why 
is he the naughtiest ? The gentleman got lost. Did the 
other one know where it was ? Yes. Why didn't he 
say ? For a lark (Fr. pour rigoler)." 

VAL (7) repeats the stories quite correctly. " Which is 
the naughtiest ? The one who got lost [who made the 
gentleman lose his way], Did he know the way, this boy, 
or did he not know it well ? Not well. Which is the 
naughtiest, the one who didn't know the way and the 
gentleman got lost, or the one who knew, but didn't tell 
the gentleman, and the gentleman didn't get lost after all ? 
The one who got lost." 

FER (8) : " Which is the naughtiest 1 The first, because 
he made the gentleman lose his way. With the second boy 
the gentleman didn't get lost." 

CAR (8) : The naughtiest is " the one who made the 
gentleman lose his way, and the less naughty boy is the one 
who didn't make the gentleman lose his way. If you had to 
punish them, which would you punish most ? The one 
who didn't know where it was. He made a mistake. But 
didn't the other one deceive the gentleman ? Yes, but 
the gentleman didn't get lost. And if the gentleman ^had 
got lost ? They -would have both been equally naughty.'' 

Here are examples of the other attitude beginning with 
an intermediate case. 


CHAP (y|) ; " Is one of the boys naughtier than the 
other ? No, they're both the same. Will the gentlemen be 
angry ? Yes, one more than the other. Which ? The one 
who didn't find his way. And if the gentlemen were to 
find the two boys again, what would happen ? One of the 
gentlemen would scold the boy [more than the other]. 
Which gentleman ? The one who didn't find his way, 
Would he be right ? No, he wouldn't be right, because the 
boy didn't know the way." 

DUR (7) : " Are they both the same, or is one naughtier 
than the other ? One is naughtier than the other. Which ? 
The one who did it for a lark. Did the gentleman get 
lost, or not IHe didn't get lost. And the other one, when 
the boy made a mistake ? $Ie did get lost. Which boy 
would you punish most ? The one who did it for a lark." 

CLAI (7) : " Which is the naughtiest ? The one who 
knew where the street was [therefore who practised deceit] 
and afterwards the gentleman didn't get lost/' 

QUEM (8|) : " Which is the naughtiest ? The one who 
knew but didn't say where it was. Did the gentleman lose 
his way ? No, the gentleman didn't lose his way because he 
asked another gentleman. And what did the gentleman in 
the other story do ? He lost his way" 

LOURD (8 ; 3) : The naughtiest " was the lie of the boy 
who knew it wasn't there. The other boy spoke without 
knowing what he was saying. Should they be punished ? 
The boy who said it wrong was tfie one who ought to have said 
it when he knew." 

KEI (10) : The naughtiest is the one " who did it on pur- 
pose. The first one deceived somebody, but he didn't know it." 

These answers confirm our findings in connection with 
clumsiness and stealing. The younger children are inclined 
to ignore the intention and to think only of the actual 
result of the action. The older ones, on the contrary, pay 
much more attention to motives. Very interesting in this 
connection is the answer given by Chap (7! yrs.), who 
knows that grown-ups too are apt to fix on the mere 
consequences of the act, but who thinks, nevertheless, that 
the grown-up is wrong and that the intention counts for 
more than the material deed. 

All these data would therefore seem to converge on one 
point. The problem we must now discuss is how the child 


ever gets beyond this moral realism and comes to judge of 
conduct by intentions. At this point a more profound 
analysis of the previous results and a few additional 
questions on the reasons adduced for not lying will put us 
on the right lines. 

it that during these early years lies seem to give rise to so 
paradoxical a realism in morality, a realism testifying to so 
faint a conception of what constitutes the real gravity of 
deceit ? How, starting from objective responsibility, will 
the child ever attain to a psychological evaluation of lying ? 
These two questions are closely bound up with each other. 

The matter can be explained quite simply if we com- 
pare our present data with the far more certain results 
obtained by observing the development of the rules of 
games. For on this last point we were able to establish 
the existence of two distinct processes in the evolution of 
children's moral judgments. On the one hand, there was 
the constraint of the adult or older child which, far from 
putting an end to egocentric thought or behaviour, easily 
combined with it and led to a purely external and realistic 
conception of rules without effective influence on the 
practice of these rules. On the other hand, there was co- 
operation, which appeared to us to dislodge both egocentric 
practice and the mystical attitude to constraint, and led 
both to a successful application of rules and to a wider 
and more interiorized understanding of what they meant. 
The following reflections therefore suggest themselves in 
connection with lies. In the first place, moral realism is 
the result of the meeting of egocentrism and constraint. 
The child, owing to his unconscious egocentrism, tends 
spontaneously to alter the truth in accordance with his 
desires and to neglect the value of veracity. The rule that 
one must not lie, imposed by adult authority, will there- 
fore seem all the more sacred in his eyes and will demand 
all the more " objective " an interpretation just because it 
does not in fact correspond with any felt inner need on his 


part. Hence, moral realism and objective responsibility, 
which stand for an inadequate practical application of the 
rule. In the second place, in so far as habits of coopera- 
tion will have convinced the child of the necessity of 
not lying, rules will become comprehensible, will become 
interiorized, and will no longer give rise to any judgments 
but those of subjective responsibility. 

We shall examine these two points in succession. In 
the case of the first we shall simply point to the conse- 
quences of our preceding remarks, while in that of the 
second we shall analyse certain fresh material contained 
in the reasons given by children for and against lying. 

Everyone knows, thanks to the fine work done by Stem 
and his followers, that until the age of 7-8 the child finds 
systematic difficulty in sticking to the truth. Without 
actually lying for the sake of lying, i.e. without attempting 
to deceive anyone, and without even being definitely 
conscious of what he is doing, he distorts reality in 
accordance with his desires and his romancing. To him a 
proposition has value less as a statement than as a wish, 
and the stories, testimony and explanations given by a 
child should be regarded as the expression of his feelings 
rather than of beliefs that may be true or false. As Stern 
has said, there are such things as apparent or pseudo-lies 
(" Pseudo - oder Scheinliige "). Since then, children's lies 
have been studied from every point of view, and every- 
where the spontaneity of pseudo-lies before the age of 7-8 
has found confirmation. 

What has perhaps been less generally understood is that 
this feature of child psychology is of an intellectual as well 
as of a moral order, and that it is connected with the laws 
of child-thought in general and with the phenomenon of 
intellectual egocentrism in particular. For the need to 
speak the truth and even to seek it for oneself is only 
conceivable in so far as the individual thinks and acts as 
one of a society, and not of any society (for it is just the 
constraining relations between superior and inferior that 
often drive the latter to prevarication) but of a society 


founded on reciprocity and mutual respect, and therefore 
on cooperation. 

For what, after all, is the spontaneous mental attitude 
of the individual ? Even before the appearance of lan- 
guage, on the motor plane which we mentioned in con- 
nection with game rules, we find that although the child's 
activity is constantly being conditioned by an ever closer 
adaptation to the things around Mm, it tends nevertheless 
to utilize these things for the sake of exercising some 
organic faculty or for the satisfaction of some psycho- 
biological tendency. In this way things, external objects, 
are assimilated to more or less ordered motor schemas, 
and this continuous assimilation of objects the child's 
own activity Is the starting point of play. Not only 
this, but when to pure movement are added language 
and imagination, the assimilation is strengthened, and 
wherever the mind feels no actual need for accom- 
modating Itself to reality, its natural tendency will be 
to distort the objects that surround it In accordance 
with its desires or its fantasy, in short to use them for its 
satisfaction. Such is the intellectual egocentrism that 
characterizes the earliest forms of child-thought. 

It is to these circumstances that we must ascribe that 
striking feature of the child's first beliefs of being im- 
mediate and not controlled (of being, as Pierre Janet says, 
merely assertive and not "reflective"). Every thought 
that enters the head of a child of 2-3 does so from the first 
in the form of a belief and not in the form of a hypothesis 
to be verified. Hence the very young child's almost 
systematic romancing as a sort of game which the 
child plays with himself as well as with others and to 
which one cannot yet give the name of pseudo-lie, so close 
is the connection between primitive romancing and 
assertive belief. Hence, finally, the pseudo-lie which is a 
sort of romancing used for other people, and serving to 
pull the child out of any strait due to circumstances, 
from which he deems it perfectly natural to extricate 
himself by inventing a story. Just as, from the intellectual 


point of view the child will elude a difficult question by 
means of an improvised myth to which he will give 
momentary credence, so from the moral point of view, an 
embarrassing situation will give rise to a pseudo-lie. Nor 
does this involve anything more than an application of 
the general laws of primitive child thought, which is 
always directed towards its own satisfaction rather than 
to objective truth. It is as his own mind comes into 
contact with others that truth will begin to acquire value 
in the child's eyes and will consequently become a moral 
demand that can be made upon him. As long as the child 
remains egocentric, truth as such will fail to interest him 
and he will see no harm in transposing facts in accordance 
with his desires. 

Thus we see that the child is almost led to tell lies or 
what seem to us as lies from our point of view by the 
very structure of his spontaneous thought. Given this 
situation, what will be the result of the laws laid down by 
adults about truthfulness ? On the occasion of the first 
very obvious lies, or of those connected with some offence 
or other and told therefore with the object of averting 
punishment or scolding, the parents point out to the child 
that he has just done something very wrong and thus 
inculcate in him the respect for truth. We agreed with 
M. Bovet in allowing that commands of this kind, laid 
down for the child by persons for whom he feels respect 
and often on the occasion of particularly strong affective 
situations, are sufficient to arouse in the child's mind 
obligations of conscience, i.e. the feeling of certain definite 
duties such as that of not telling lies again. Moreover, as 
we saw in the game of marbles, a rule may be felt as 
sacred and obligatory without, for that matter, being 
properly applied. One may even say that in certain cases 
the more defective the application of a rule the more the 
rule will be felt as obligatory, given the continual conflicts 
(and consequently the feelings of guilt) which these 
infractions of the rule must lead to. Be that as it may, we 
can say that up till the age of 7-8, the child tends spon- 


taneousiy to alter the truth, that this seems to him 
perfectly natural and completely harmless, but that he 
considers It a duty towards the adult not to lie, and 
recognizes that a lie is a " naughty " action. Moral 
realism and objective responsibility are the inevitable 
outcome of so paradoxical a situation. 

Two groups of causes conspire, it is true, to bring about 
this result. The first group is of a general order and will 
be dealt with at the conclusion of this chapter : the child 
is a realist in every domain of thought, and it is therefore 
natural that in the moral sphere he should lay more stress 
on the external, tangible element than on the hidden 
motive. The second group of causes is peculiar to the 
situation we have been describing. It is obvious that if 
the desire for truthfulness does not correspond to some- 
thing very fundamental in the child's nature, the adult's 
command, in spite of the nimbus that surrounds it, will 
always remain external, " stuck on " as it were, to a 
mind whose structure is of a different order. For the 
spirit of such a command could only be understood by 
experience. One must have felt a real desire to exchange 
thoughts with others in order to discover all that a lie can 
involve. And this interchange of thoughts is from the 
first not possible between adults and children, because the 
initial inequality is too great and the child tries to imitate 
the adult and at the same time to protect himself against 
Mm rather than really to exchange thoughts with him. 
The situation we have described is thus almost the neces- 
sary outcome of unilateral respect. The spirit of the 
command having failed to be assimilated, the letter alone 
remains. Hence the phenomena we have been observing. 
The child thinks of a lie as " what isn't true ", inde- 
pendently of the subject's intentions. He even goes so far 
as to compare lies to those linguistic taboos, " naughty 
words ". As for the judgment of responsibility, the 
further a lie is removed from reality, the more serious is 
the offence. Objective responsibility is thus the inevitable 
result of unilateral respect in its earliest stage. 


Let us now turn to the second problem we set ourselves 
and let us ask how the child is going to acquire a real 
understanding of lying and become capable of judging of 
subjective responsibility. It follows quite naturally from 
what we have said that the passage from unilateral 
respect to mutual respect is what liberates the child from 
his moral realism. But before carrying the discussion any 
further, let us pause to examine certain facts that are very 
interesting in this connection. 

For if what we have said is true, then two questions are 
forced upon us. The first is : what are the child's ideas 
about the moral utility of not lying ? The next is : from 
what moment and in what conditions does the child 
consider it a moral fault to lie to his equals ? And we 
shall see that the answers given to these two kinds of 
question go to show that in mutual respect and coopera- 
tion is to be found the real factor that brings about a 
progressive understanding of the law of truthfulness. 

Let us first try to see why people should not tell lies. 
The reason most universally invoked and that which 
comes first chronologically is that you mustn't tell lies 
because " you get punished ". Here are some examples : 

ZAMB (6) : " Why must we not tell lies ? Because God 
punishes them. And if God didn't punish them ? Then 
we could tell them." 

Roc (7) : " What happens when you tell lies ? You get 
punished. And if you didn't get punished, would it be 
naughty to tell them ? No. I'm going to tell you two 
stories. There were two kiddies and they broke a cup 
each. The first one says it wasn't him. His mother 
believes him and doesn't punish him. The second one also 
says that it wasn't him. But his mother doesn't believe 
him and punishes him. Are both lies that they told 
equally naughty ? No. Which is the naughtiest ? The 
one who was punished." 

BURD (7) tells- us, in connection with the stories of the 
dog and the school teacher, as we saw in the preceding 
paragraph, that the lie of which the mother knows that it 
is untrue is the worst. Then he adds spontaneously the 
following curious explanation : " What she knows is the 


worst. Why ? Because she she can scold 

at once. When she doesn't she can't scold at once. 

The child doesn't he is being scolded. He doesn't 

remember any more. But when is a child naughtiest, when 
he is scolded at once or when he is not scolded at once ? 
He is naughtier when he is scolded at once"" 

DELE (9) : st Why is it naughty ? -Because you gd 
punished. If telling lies wasn't punished would it be 
naughty ? Oh, no 1 " 

The following observation recently made at the Maison 
des Petite by Mile Lafendel is worth recording : A little 
girl of 6 had just told a lie. Mile L. asked her if that was 
right. The answer was : " It doesn't matter, my mummy 
can't see ! 9> 

Note how closely observed is the situation as described 
by Burd. But what an extraordinary morality ordinary 
education produces the child is all the naughtier for 
being scolded straight away ! These children, in short, 
look upon lying as naughty because it is punished, and if 
it were not punished no guilt would attach to it. This is 
objective responsibility in its purest form. These facts, 
moreover, should not be interpreted as even a relative 
amorality. The child does not mean that it is enough to 
escape censure to be innocent. What these subjects think 
is simply that the punishment is the criterion of the 
gravity of the lie. Lies are forbidden, though one does 
not quite know why. The proof is that you get punished 
for it. If it were not punished, it would not be " naughty ". 
What these answers mean, then, is that a lie is a fault in 
so far as it is forbidden by God or adults. This is heter- 
onomy in its most naive form and it confirms our interpre- 
tation of realism in its beginnings. 

Here are some slightly more advanced answers. A lie 
is a fault in itself, and would remain so even if it were not 

DUR (7) : " Are we allowed to tell lies ? No. It's 
naughty. Why ? Because we get punished. If we could 
tell them without being punished, would it be naughty or 
not naughty ? Naughty. Supposing it was in a country 
where there were only children and where no one would 


know If we told lies, would it still be naughty to tell them 
or not ? It would be naughty." 

GIR (9) : " Why is it naughty ? Because we get pun- 
ished. If you didn't know you had told a lie, would it be 
naughty too ? It would be naughty, but less naughty [we 
have here an interesting residue of the last type of answer]. 
Why would it be naughty ? Because it is a lie all the 

AUF (9) : " If lies were not punished, would they still be 
wrong ? They would still be wrong for the boy who said 

ARL (10) : " If lies were not punished, would they be 
naughty or not ? Of course they would be naughty" 

At a certain stage, then, rules become obligatory inde- 
pendently of punishments, that is to say, independently of 
the controlling power whence they emanate. This is the 
phenomenon to which P. Bovet has drawn attention. 
Commands, which are at first closely bound up with the 
person who imposes them, are later on elaborated by the 
child's reason and through this become universal. It is a 
process of the same order as that described by P. Janet 
in connection with his " rational " stage. The laws of 
conduct, like those of reflective thought, end by being 
placed above the actual context in which they are ex- 
perienced, and thus become universal and absolute. 

But even though they are generalized in this way, rules 
are none the less heteronomous. The child, it is true, 
takes the particular command that has been given him 
and raises it to the level of a universal law. This rational 
process of extension is probably already due to coopera- 
tion. But the rule may nevertheless persist in the form of 
an imperative that is external to the child's own con- 
science. We shall now see how this state of things is left 
behind at a given moment, thanks to the growing under- 
standing of the child himself. 

In point of fact, the older children of 10-12 generally 
invoke against lying reasons which amount to this : that 
truthfulness is necessary to reciprocity and mutual agree- 
ment. Among the alleged motives there will be found, it 
is true, a whole set of phrases inspired by adult talk : 


" We mustn't tell lies because It's of no use Sl (Arl, 10). 

" We must speak the truth . . . our conscience tells us 
to" (Hoff, ii). But along with these commendable 
but too often meaningless formulae we can observe a 
reaction which seems to be, if not altogether spontaneous, 
at any rate founded on experience. The reaction in 
question implies that truthfulness is necessary because 
deceiving others destroys mutual trust. One is struck by 
the fact, in this connection, that while the younger chil- 
dren had regarded a lie as all the worse for being un- 
believable, the older ones, on the contrary, condemn a lie 
in so far as it succeeds. 

Here are some fresh examples of the attitude of the 
younger children on this point, which, incidentally, we had 
occasion to notice more than once during the preceding 

BUG (6) : The lie about the dog is worse than the one 
about the good marks " because his mother would have 
believed " the latter. 

Roc (7) : Lies are " not nice. Why not ? Because 
sometimes you believe them and sometimes you don't. 
Which are worse, those you believe or those you don't 
believe ? Those you don't believe, Two children both say 
they have sore feet so that they won't be sent messages. 
One is believed, the other not. Which one is naughtiest ? 
The one you didn't believe. Why ? Because you can 
make a mistake by looking [ = because in the case of the 
child who was believed, appearances might have let you 
think that he had a sore foot : the lie was probable and 
therefore less bad than that which had appearances too 
much against it]." 

SAV (7) : " When is a lie worse, when people believe it or 
when they don't believe it ? When you can't believe in it." 

The older children, on the contrary, hold that a lie is 
"bad precisely in so far as it achieves its aim and succeeds 
in deceiving the other person. Those, therefore, who have 
really grasped the anti-social character of lying no longer 
say that we mustn't lie " because we get punished ", but 
because to do so is contrary to reciprocity and to mutual 


DIN (8, forward) : " Why must we not lie ? -Became if 
everyone lied no one would know where they were!' 

AUF (9) : " Why is it naughty to tell a lie ? Became 
mother believes it!" Exactly the opposite criterion from 
that used by the little ones. 

Loc (10) : " Why is it naughty to tell a Me ? Because 
you can't trust people any more." 

KEI (10) : The lie about the school teacher is worse 
than the lie about the dog, because "the^one about the 
teacher you couldn't know [that it was a lie]." 

It would seem, then, that the evolution of the answers with 
age marks a definite progress in the direction of reciprocity. 
Unilateral respect, the source of the absolute command, 
taken literally, yields the place to mutual respect, the source 
of moral understanding. We can, indeed, distinguish three 
stages in this progress. In the first stage, a lie is wrong 
because it is an object of punishment ; if the punishment 
were removed, it would be allowed. Then a lie becomes 
something that is wrong in itself and would remain so even 
if the punishment were removed. Finally, a lie is wrong 
because it is in conflict with mutual trust and affection. 
Thus the consciousness of lying gradually becomes interior- 
ized and the hypothesis may be hazarded that it does so 
under the influence of cooperation. If we attribute the 
advance to the child's intelligence alone, which is constantly 
improving his understanding of what he originally took 
in a purely realistic sense, we are only shifting the 
question. For how does psychological intelligence advance 
with age if not by means of increased cooperation ? 
Cooperation, of course, presupposes intelligence, but this 
circular relation is perfectly natural : Intelligence animates 
cooperation and yet needs this social instrument for its 
own formation. 

If our hypothesis is correct, then we ought now to find 
and this is the second question we have to examine 
that lies between children, which are at first held to be 
legitimate, 6nd by being proscribed from among the 
relations they hold with each other. And this indeed does 
prove to be the case. 


Here, to begin with, are some answers given by the 
younger children, who that lies between children are 

PEL (6) : " Is It just the same to tell lies to grown-ups 
and to children ? A T o. Which is naughtiest ? To grown- 
ups. Why ? Because they know it isn't true. [Note the 
spontaneous reappearance of the realistic criterion: the 
more a lie shows, the worse it is,] And how about 
children ? It's allowed because they are smaller.'* 

BLI (6) ; * f To a grown-up it's not ike same thing as to 
children. Which is naughtiest ? To grown-ups.'* 

IRI (7) : " And may you tell lies to other kiddies or 
not ? Yes, M'sieu. Is it naughty or not ? It's a little 
bit naughty too. Is it worse to grown-ups, or to other 
kiddies, or are they both the same ? It's not so naughty 
to children. Why ? Because they are not big.' 3 

DRUS (7) : " It's naughtier to a grown-up. They are 
bigger than children, and little children can say naughty 
words to each other [cf. the definition of a lie !]." 

CARN (8) : You can tell lies to children but not to 
grown-ups " because a gentleman is worth more than a child." 

EM (8 ;. 5) : "A child doesn't know if it is true. A 
grown-up knows, so it's naughtier ! " Cf. the realistic 
criterion rising spontaneously in a child who has never 
been questioned by us before. 

Pi (9 ; 3) : " It's naughtier to a grown-up because they 
are older/' 

ER (9 ; 8) : " It doesn't matter to a child. You can tell 
him lies. But you mustn't to a grown-up." 

Almost all the older children, on the contrary, though 
they maintain that it is wrong to deceive adults, think 
that it is as bad or even worse to deceive one's comrades. 
You can say things for a joke, but a serious lie is as 
reprehensible between children as towards an adult. 

Here are examples. 

BOH (7 \ 10) : " Can one tell children lies, or is that just 
as bad as telling them to grown-ups ? They are both the 
same thing.*' 

Di (8 ; 6) : " It's just as bad to a child, because a child 
can't tell if you're lying or not, and he's not pleased." The 
argument is exactly the opposite of Em's. 

AUD (9!) : " It is worse to a child, because a child is 


DEN (n yrs.) : " It's worse to a child because he'll 
believe it" 

COTT (12 ; 8) : " You get into more of a row with a 
grown-up, but they're just as bad as each other." 

CAL (12 yrs.) : " Sometimes you almost have to tell lies to 
a grown-up, but it's rotten to do it to another fellow." 

These answers I show how much the progress of soli- 
darity between the children helps them to understand the 
true nature of lies. Compare, for example, Em's answers 
with those given by Di. To Em, who is dominated by 
authority, a lie is wrong in so far as it show r s, but to Di, 
who has learned comradeship, a lie is wrong in so far as it 

In conclusion, we can solve as follows the two questions 
we set ourselves above. If the younger children present 
all the features of an almost systematic moral realism, 
leading in certain cases to the predominance of objective 
over subjective responsibility, this is because of the sui 
generis relations holding between adult constraint and 
childish egocentrism. The child's unilateral respect for 
the adult obliges him to accept the latter 's commands, 
even when these are not such as can be put immediately 
into practice. Hence the externality of rules and the 
literal character of the moral judgments to which they 
give rise. If, conversely, the child's development is such 
as to interiorize these commands and to make his own the 
notion of subjective responsibility, this is because coopera- 
tion and mutual respect are giving him an increasingly 
vivid understanding of psychological and moral realities. 
Thus truthfulness gradually ceases to be a duty imposed 
by heteronomy and becomes an object envisaged as good 
by an autonomous personal conscience. 

In short, the study of rules imposed upon the child by 
the adult broadly confirms what we learnt from the study 
of rules which younger children receive from older ones. 
However closely connected mutual and unilateral respect 
may be in the general continuity of mental phenomena, 

1 See statistics, p. 308. 


the two processes lead nevertheless to consequences that 
are qualitatively different. 

5. CONCLUSION. MORAL REALISM. We must now try 
to find what place our results occupy in the moral life of 
the child taken as a whole by restoring things to the right 
perspective which has inevitably suffered distortion 
through the method of interrogation. 

Two distinct levels of activity are to be distinguished 
in moral thought. First, there is effective moral 
thought, t moral experience " which is built up gradu- 
ally in action as the subject comes in contact with 
reality and meets with shocks and opposition. It is that 
which leads him to form such moral judgments as will 
guide him in each particular case as it comes Ms way and 
enable him to evaluate other people's actions when these 
concern him more or less directly. And there is also 
theoretical or verbal moral thought, bound to the former 
by all kinds of links, but as far removed from it as is 
reflective thought from immediate action. This verbal 
morality appears whenever the child is called upon to 
judge other people's actions that do not interest Mm 
directly or to give voice to general principles regarding his 
own conduct independently of his actual deeds. 

The analysis we were able to make of the judgments on 
responsibility deals only with the cMld's theoretical moral 
thought, and is in no way concerned with Ms practical 
and concrete moral thought (differing in this from our 
enquiry on the rules of a game where we were able to 
keep both aspects of the question simultaneously in view). 
Now with regard to this verbal plane our results were 
fairly consistent. Though we could not point to any 
stages properly so called, wMch followed one another 
in a necessary order, we were able to define processes 
whose final terms were quite distinct from one another. 
These processes might mingle and overlap more or less in 
the life of each child, but they marked nevertheless the 
broad divisions of moral development. We saw, for 


example, that the child's theoretical morality could be 
_ subject either to the principles arising from unilateral 
respect (morality of heteronomy and objective responsi- 
bility) or to those based on mutual respect (morality of 
inwardness and subjective responsibility). 

But the problem before us now and already touched 
upon in the methodological remarks at the opening of the 
present chapter is this. What do these results correspond 
to in the child's effective thought on morality ? Two 
solutions are possible. It may be that verbal thought is a 
progressive conscious realization of concrete thought. In 
that case the moral realism we met with and studied 
would correspond to a moral realism effectively at work in 
action, a realism that would no doubt already have been 
left behind by the time the children talked about it 
theoretically, but a realism which would none the less give 
rise to spontaneous reactions expressed in concrete acts. 
But it may also be that this verbal morality, whose 
manifestations we have observed, corresponds to nothing 
at all in the child's effective thought. The children would 
therefore never have manifested any moral realism in 
their concrete decisions and judgments. Meeting in their 
lives with lies analogous to those of the dog as big as the 
cow or of the boy who pretended to have had good marks 
in order to be rewarded, they would never have hesitated 
to regard the second as worse than the first. And, in this 
case, verbal reflection alone would be sufficient to engender 
a form of pure psittacism bearing no relation to past or 
present action. 

First let us remove a possible ambiguity. This second 
solution is not equivalent to saying that the interrogatory 
is the cause of the whole trouble. In life, the child is 
constantly finding himself in situations analogous to those 
described in the interrogatory. He hears his friends' 
misdeeds discussed, he has the chance of judging them 
from hearsay, and phenomena similar to those we have 
observed may very well occur in a perfectly spontaneous 
manner. In the same way, the results concerning animism 


or artificialism which we obtained earlier (see C.W,) 
correspond to remarks that can frequently be heard 
from the lips of children. But this is only shifting the 
problem a stage further off, and we still have to ask 
concerning these spontaneous verbal reflections : do they 
or do they not constitute the comiBg into consciousness 
of some effective form of thinking ? 

As a matter of fact, it is our belief that even for the 
child, theoretic moral reflection does constitute a pro- 
gressive conscious realization of moral activity properly so- 
called. Consequently, we think that the results set out above 
correspond in a certain measure to real moral facts. But 
the relations between thought and action are very far from 
being as simple as is commonly supposed, and it will there- 
fore be necessary to stress somewhat the point under discus- 
sion in order to grasp our results in their true perspective. 

In the first place, as Claparede has shown so well, ideas 
or notions enter into consciousness in inverse order to 
that in which they actually arise : what comes first in the 
order of action comes last in the order of conscious realiza- 
tion (prise de conscience). So that if on the verbal plane 
moral realism seems to be a primitive fact, it does not follow 
that it is such on the plane of action itself. The notion 
of good which, generally speaking, appears later than 
the notion of pure duty, particularly in the case of 
the child, is perhaps the final conscious realization of 
something that is the primary condition of the moral 
life the need for reciprocal affection. And since moral 
realism is, on the contrary, the result of constraint 
exercised by the adult on the child, it may perhaps be a 
secondary growth in comparison to the simple aspiration 
after good, while still remaining the first notion to be 
consciously realized when the child begins to reflect upon 
morality and to attempt formulation. 

In the second place, to realize consciously is not simply 
to throw into light ideas that have already been fully 
worked out. Conscious realization is a reconstruction and 
consequently a new and original construction superimposed 


upon the constructions already formed by action. As 
such, therefore, it comes on the scene later than action 
proper. Hence the time-lag noted by Stem, and which 
we have found in every sphere of child thought l . If, 
then, the moral realism we have noted between 6 and 8 
on the average (we could not detect it under 6 because 
the youngest children could not sufficiently understand the 
stories used), if, then, this moral realism corresponds to 
something present in the moral activity itself, it is not 
during those years that we must seek for this something, 
but at a very much earlier period. For objective responsi- 
bility can perfectly well have been discarded long ago on 
the level of action and yet subsist on the level of theo- 
retical thought. Besides, we saw examples of children 
who judged the stories we told them in accordance with 
the principles of objective responsibility, but who at the 
same time gave us personal reminiscences evaluated in com- 
plete conformity with the criterion of subjective responsi- 
bility. This being so, can we find during the first years 
of moral development instances of realism and objective 
responsibility that correspond to the phenomena which 
we observed on the verbal plane ? We believe that we can. 
In the first place it should be noted that, however 
averse one may be in education to the use of any con- 
straint, even moral, it is not possible completely to avoid 
giving the child commands that are incomprehensible to 
it. In such cases which are almost the rule in the 
traditional form of education based on authority the 
mere fact of accepting the cotnmand almost invari- 
ably provokes the appearance of moral realism. Here 
are a few examples observed on one of the author's own 
children, on a child, therefore, whose parents have done all 
they could to avoid objective responsibility. The following 
observations will thus hold a fortiori for children in whose 
immediate entourage no special attention is devoted to 
this complex problem of moral autonomy and heteronomy. 

1 See Bull. Soc. Franpaise et Philos., 1928, pp. 97 ei seq. t especially 
p- 105. 


Jacqueline has never been punished in the strict sense 
of the term. At the worst, when she makes a scene, we 
leave her alone for a little while and tel her we shall come 
back when she can talk quietly again. She has never 
been given duties as such, nor have we ever demanded 
from her that sort of passive obedience without discussion 
which in the eyes of so many parents constitutes the 
highest virtue. We have always tried to make her under- 
stand the ** why " of orders instead of laying down 
" categorical " rules. Above all, we have always put things 
to her in the light of cooperation : " to help mummy ", 
to " please " her parents, to ee show her sister ", etc. are 
for her reasons for carrying out orders that cannot be 
understood in themselves. As to rules that are unin- 
telligible to very little children, such as the rale of 
truthfulness, she has never even heard mention of them. 

But in ordinary life it is impossible to avoid certain 
injunctions of which the purport does not immediately 
seem to have any sense from the child's point of view. 
Such are going to bed and haying meals at given hours, 
not spoiling things, not touching the things on daddy's 
table, etc. Now, these commandments, received and 
applied before being really understood, naturally give rise 
to a whole ethic of heteronomy with a feeling of pure 
obligation, with remorse in case of violation of the law, etc. 

For example, one evening I find Jacqueline, aged 2 ; 6 
(15) a , in bed, spoiling a towel by pulling out the threads 
one by one. Her mother has already often told her that 
it is a pity to do that, that it makes holes, that you can't 
mend the holes, etc. So I say to J. : " Oh, but mummy 
will be sad/' J. answers calmly and even with an ill- 
concealed smile : " Yes. It makes holes. You can't 
mend "... etc. I continue my lecture, but she obvi- 
ously is not going to take me seriously. Still hiding her 
amusement with difficulty, she suddenly says to me 
" Laugh ! " in so comic a tone, that in order to keep a 
straight face I quickly change the subject. J., very 
conscious of her powers of seduction, then says to me " My 
little darling Daddy ", and the incident ends. The next 
morning, however, J. wakes up full of it. Her first words 
refer to what had happened the night before. She thinks 
about the towel and asks her mother whether she isn't sad. 
So in spite of the first reaction showing such charming 

1 2; 6 (15) =2 years, 6 months and 15 days. 


disrespect, my words had told and the command had 
brought about the usual consequences. 

The evening of the same day, J. begins to puU the 
threads out of the towel again. Her mother repeats that 
it is a pity. ]. listens attentively but says nothing. A 
moment later she is calling out and cries till someone 
comes to her : she simply wanted to see her parents again 
and make sure that they bore her no grudge. 

We have here an example of a command bringing about, 
with or without apparent respect, a well-marked feeling of 
duty and of wrong. Now it seems obvious to us that 
such feelings are set up before the child has any clear 
consciousness of moral intention, or at any rate before it 
can distinguish between what is " done on purpose " (an 
action carried out knowingly and in voluntary defiance of 
the command) and what is " not done on purpose ". A 
child of two and a half spoiling a bath towel has obviously 
no intention of doing harm. It is simply making an 
experiment in physics. Even if you ask it to stop, it may 
forget the command and begin again, or remember the 
rule too late to resist the first impulse. We would there- 
fore not class J/s conduct as described above among the 
acts of disobedience (which are by definition conscious) 
nor a fortiori among those committed with the intention 
of doing harm. And yet the sense of guilt is clearly 
present. It was not till after she was three years old that 
we noted in J. any reactions implying the notion that 
something " not done on purpose " could not be reckoned 
a fault. For instance, I say to her : " You know 
you're disturbing me just a little." Answer : " J didn't 
think " (in other words, " I didn't do it on purpose ") *. 
After the age of 4 months J., it is true, uses expressions 
like " but it's your fault ". But these apply only to the 
material results and not to the intention. And it is just 

1 It is interesting to note that the moment when the idea of intention 
appears in the moral language of 'the child coincides more or less with 
the age of the first " whys ". Indeed, as we have tried to show else- 
where (L.T,, chap. V), that the first " whys " correspond precisely to a 
need for motivation which results from the conscious realization of 
the intentionality of actions. 


at this age, from the time when speech begins up to about 
three, that It is easiest to observe moral realism and 
objective responsibility in their pure and spontaneous 

Let us turn, therefore, to obvious cases of objective 
responsibility. In the earliest stages of life cleanliness is a 
constant occasion for judgments of responsibility. Now it 
goes without saying that in such a domain the element of 
intentionality is almost entirely absent from the children's 
actions. Even the most orthodox Freudians cannot deny 
"that in the vast majority of cases the child of one or two 
that cannot control its bodily functions when it is asleep 
or at play has simply not acquired the necessary physical 
control. Cases of intentional or even " unconscious " 
resistance to the rules of cleanliness are very rare at this 
age. Now, in spite of the purely automatic character of 
the reactions in question, parents are bound to ask their 
children to be careful. When accidents occur it is only 
natural that they should express their disappointment. 
In short, however delicately one may put the matter, 
there have to be commands and therefore duties. Now 
the curious thing is that the sense of guilt is proportional, 
not to the incidental negligence (as when the child forgets 
to " ask ", etc.), but to the physical acts themselves. 

For example, J. at I ; n (28) has been put to bed and 
given medicine of whose effects she has been informed. 
In spite of her mother's precautions, specially designed to 
avoid any reaction of shame or guilt, J. is greatly upset 
when the medicine works. Her face assumes an expression 
of distress, her eyes fill with tears, her mouth droops, and 
she is obviously experiencing the same feelings as if the 
thing had happened in normal circumstances through her 
own negligence. 

But it must not be thought that this special domain is 
the only one propitious to the spontaneous development 
of objective responsibility. Any rule involving a material 
application admits of the same deviations. Another good 
example is that concerning food and the rules connected 
with it. 


For some time J, has had a very small appetite, with 
the result that during this period of her life the essential 
rules of her universe were those appertaining to food. 
The World-Order decreed that one should take a cup of 
cocoa at four o'clock, a good bowlful of vegetables in the 
middle of the day, a few little drops (of hydrochloric 
acid) in water just before lunch, etc. Now once these 
orders had been accepted, right and wrong were defined 
by the conformity or non-conformity of actions in ^ re- 
lation to them, and this independently of all possible 
intentions or circumstances. For example, one day J. at 
2 ; 10 (7) is not very well and her mother feels that 
probably the usual plate of vegetables will be too much 
for her. Sure enough, after one or two mouthfuls^ J. 
shows signs of weariness. But she insists upon finishing 
her helping, because it is the rule. It is no good letting 
her off, she perseveres in her view, though she is not 
enjoying her food. Every time she is given a spoonful she 
cannot swallow it, but when the bowl is taken away she 
asks for it back, as though it were a sin not to empty it. 
Finally it is taken away and we try to reassure her by 
telling her that it is not her fault, that some days people 
are less hungry than others, etc. In spite of aU these 
precautions taken by her mother, J. then begins to cry. 
Even when she has been comforted she still shows signs of 
remorse, promises to go to sleep, etc. 

Another example. At the age of 2 ; 10 (23). J. is taking 
her hydrochloric acid as usual. But too many drops have 
been put in the glass, and J. is told that she need not 
drink it all. Sure enough, after taking a draught or two 
she complains that it prickles ; she looks disgusted and 
even feels sick. AU the same she wants to drink it all up. 
Her mother repeats that it is not necessary and lifts her 
down from her chair. J. bursts into tears as though she 
had done wrong. She comes back to the glass and insists 
upon drinking it up. 

These last two examples seem to demonstrate how 
strong and spontaneous is the child's evaluation of ob- 
jective responsibility. It is even staggering to find that in 
a little girl who has never known what authority is and 
whose parents make a point of cultivating autonomy of 
conscience in their children, the orders received should 
lead to so stubborn a moral realism. A rale emanating 


from the parents brings about a sense of duty against which 
the later attenuations of the parents themselves are for the 
moment powerless. It Is true that in the three cases we have 
just quoted (cleanliness and rules about food) pride may 
play a certain part. The child will not admit defeat. But 
this very pride presupposes a realistic consciousness of rules. 
If the child did not consider it a sort of moral lapse in 
itself not to finish up her glass of medicine, her bowl of 
vegetables, or her cup of cocoa, she would not feel humili- 
ated at being let off these obligations 1 . 

A third group of examples relating to clumsiness must 
also be given. This will bring us back to the facts we have 
studied in the course of this chapter, since what we sought 
to analyse was precisely how the child evaluates the acts 
of clumsiness that he hears about in the set stories. Now, 
during the early years, a child will often regard Ms own 
clumsiness from a purely objective point of view, even to 
his own detriment. 

J., at about 2 years old, is playing with a shell I have 
lent her. The shell is very fragile and breaks the first 
time it is dropped. J. is dismayed, and I have the 
greatest difficulty in persuading her that it is not her 

We need not dwell upon this example. It is a common- 
place among children, and everyone will have had occasion 
to observe something like it. On this point too it is not 
until later that the child will differentiate involuntary 
clumsiness (pure accident) from clumsiness due to negli- 
gence or carelessness. In the early stages, it is only the 
result that counts. 

In short, it is our belief that during the first years, the 
inevitable constraint of the adult even if, as in the case 
of J., it is reduced to the minimum necessarily brings 
about a certain moral realism which will be more or less 
marked according to the nature of the home and to the 

1 We may add, In order to reassure the reader, that Jacqueline is 
by no means continually haunted by the commands which she respects . 


combined characters of parents and child. The moral 
realism which later on we observe on the verbal plane 
would thus be the indirect result of these early phenomena. 

But between the spontaneous moral realism of the early 
years and the theoretical moral realism which we analysed 
before, there is an intermediate link that must not be 
disregarded we mean the judgment made by the child* 
not about his own actions, but about the conduct of his 
equals. As far as he himself is concerned he succeeds 
fairly soon (at about 3-4, when the first " whys " and the 
interest in motivation begin) in differentiating intentional 
faults from involuntary breaches of the moral code. And 
soon after this he learns to excuse himself by the plea pf 
" not on purpose ". But when it comes to the deeds of 
those around him, things appear in a very different light. 
Generally speaking, it is not going too far to say that the 
child like ourselves is more severe with others than 
with himself. The reason for this is quite simple. The 
conduct of other people appears in its outward shape long 
before we can understand the intentions behind it ; so that 
we are apt immediately to compare this outward shape 
with the established rule and to judge, the action by this 
essentially objective criterion. It is only by a continuous 
effort of generosity and sympathy that we can resist such 
a tendency and try to understand other people's reactions 
in terms of their intentions. It is obvious that the child 
is capable very early of such intropathy. But it is also 
obvious that during this phase where respect for rules still 
outweighs cooperation (the stage we have called " ego- 
centric " in the game of marbles and in which rules are at 
once mystically accepted and egocentrically practised) it 
is also obvious, we repeat, that to judge psychologically 
will require a greater effort in the case of other people's 
actions than in that of our own. To put it differently, 
moral realism will last longer with regard to the evalua- 
tion of other people's conduct than with regard to that of 
one's own. And the facts, indeed, seem to bear this out. 

Here from amongst many others is an observation 


which one of our students, Mme Weissfdler, has been 
kind enough to communicate to us. Parallels to it could 
be found in any child. 

L MAD (4 yrs.) says to her mother : " Yon Mummy, 

I've wiped the dishes" Ja (2| yrs.) adds ; " Me too.'* 
Mad : *' No, that's not true. Yes. No, you didn't wipe 
the dishes. So it*s naughty to say that." Mother : " Per- 
haps Ja is making a mistake and thinks she has wiped 
them/' Mad : " No, no, she didn't them. It's net 

nice of her to say thai. It was only me.** 

II. Ja : et You're all over spots, you are. Mad : No. 
Ifs not true, it's not true, It is true. Mad (furiously) : 
It's not true. That's nonsense. . . . Mummy, Ja is talking 
nonsense ! " 

III. Ja, without meaning to, hurts Mad. Mad cries : 
" Mummy, Ja is horrid. She hit me. But she didn't do it on 
purpose, she didn't want to hurt you, Yes she did, she did 
it on purpose. She hit me very hard. She hurt me very much.'* 

In dealing with the evaluation of the conduct of others we 
have approached the slightly artificial situation in which the 
child considers actions that are not directly observed but 
are described by means of a story. If moral realism lasts 
. longer with regard to judgments made about others than 
in the purely individual domain, it goes without saying 
that its reign will be even further prolonged when it 
comes to the purely verbal examples contained in our 
stories. But here a third phenomenon comes in to compli- 
cate matters. 

For the child's moral realism is certainly far more 
systematic on the plane of theory than on that of action, 
so that we would seem to be dealing here with something 
in the nature of a new and different phenomenon. The 
fact is that the conscious realization implied in all theo- 
retical reflection does not merely repeat after a greater or 
lesser interval, what has effectively taken place in practical 
actions. There are, over and above the delays, distortions 
inherent in the very mechanism of reflection. For as soon 
as on immediate action there follows, or is superposed, a 
thought that is detached from reality and set free by the 
power of words or imagination, the mind is thrown out of 


focus, a whole set of illusions of perspective prevails, and 
especially the unconscious illusion of egocentrism. Thus it 
is that in the inteliectnal sphere, the child who tries to 
reason will be up against a series of difficulties which were 
conquered long ago by practical intelligence. Similarly, in 
the moral sphere when he simply has stories told to him, 
he will be led to make judgments devoid of pity and lacking 
in psychological insight, testifying therefore to a more or 
less systematic moral realism, whereas in real life he would 
undoubtedly sympathize with those whom from afar he 
regards as the greatest sinners. 

We see, therefore, how the spontaneous moral realism of 
the early years, while it dwindles progressively with regard 
to the subject's own conduct, may very well develop 
elsewhere, first in the evaluation of other people's actions, 
and finally in reflection concerning purely theoretical cases 
involved in stories, in histories, and in social myths in 
general. If we were dealing with so-called primitive 
societies, we should have to add that these final products 
of moral realism, once they are consolidated by the social 
constraint of the group as a whole (in contrast to the 
elementary constraint of adults over children) are capable 
of reacting upon the actual minds of individuals by a reper- 
cussion that will be readily understood. But as we are 
speaking only of the child we may confine ourselves to the 
conclusion that moral realism does correspond to some- 
thing effective and spontaneous in child thought. This 
result, moreover, is in complete agreement with what we 
found in the case of the game of marbles. Every rule, 
whether it be imposed upon the younger by the older 
child, or upon the child by the adult, begins by remaining 
external to the mind before it comes to be really interiorized. 
During this purely external phase, the most rigorous 
moral realism may very well go hand in hand with what 
seems to be the laxest and most egocentric practice. 

Having cleared up this point let us now try to solve the 
problem of the nature of moral realism taken dynamically 
and reduced to the true proportions indicated by the 


preceding discussion. In other words, let us not any 
longer regard the results of our interrogatories as self- 
contained, but rather as the and indirect outcome 
of a primitive and far more tendency. It is this 
spontaneous moral realism of which the children's theo- 
retical talk is only the reflection which we must now 
examine so as to establish its origin and conditions. 

Moral realism seems to us to be due to the conjunction 
of two series of causes those peculiar to the spontaneous 
thought of the child (childish "realism 1 '), and those 
belonging to the constraint exercised by the adult. But 
this conjunction, far from being accidental, seems to us to 
be characteristic of the most general processes of cMld 
psychology as they occur in the intellectual as well as in 
the moral domain. For the fundamental fact of human 
psychology is that society, instead of remaining almost 
entirely inside the individual organism as in the case of 
animals prompted by their instincts, becomes crystallized 
almost entirely outside the individuals. In other words, 
social rules, as Durkheim has so powerfully shown, 
whether they be linguistic, moral, religious, or legal, etc., 
cannot be constituted, transmitted, or preserved by means 
of an internal biological heredity, but only through the 
external pressure exercised by individuals upon each other. 
To put it in yet another way. As Bovet has demonstrated 
in the field of morals, rules do not appear in the mind of the 
child as innate facts, but as facts that are transmitted to him 
by his seniors, and to which from his tenderest years he has 
to conform by means of a sui generis form of adaptation. 
This, of course, does not prevent some rules from containing 
more than others an element of rationality, thus corre- 
sponding to the deepest functional constants of human 
nature. But whether they be rational or simply a matter 
of usage and consensus of opinion, rules imposed on the 
childish mind by adult constraint do begin by presenting 
a more or less uniform character of exteriority and sheer 
authority. So that instead of passing smoothly from an 
early individualism (the "social" element of the first 


months is only biologically social, so to speak, inside the 
individual, and therefore individualistic) to a state of 
progressive cooperation, the child is from his first year 
onwards in the grip of a coercive education which goes 
straight on and ends by producing what Claparede l has 
so happily called a veritable " short-circuit ". 

As a result of this we have three processes to consider : 
the spontaneous and unconscious egocentrism belonging 
to the individual as such, adult constraint, and coopera- 
tion. But and this .is the essential point the spon- 
taneous egocentrism of the child, and the constraint of the 
adult, far from being each other's antitheses on all points, 
so far agree in certain domains as to give rise to para- 
doxical and singularly stable compromises. For coopera- 
tion alone can shake the child out of its initial state of 
unconscious egocentrism;* whereas constraint acts quite 
differently and strengthens egocentric features (at any 
rate on certain points) until such time as coopera- 
tion delivers the child both from egocentrism and from 
the results of this constraint. We shall attempt to verify 
these statements with regard to moral realism, after which 
we shall compare this phenomenon with the precisely 
parallel processes that present themselves in the domain 
of child intelligence. 

The first group of factors that tend to explain moral 
realism is therefore based on one of the most spontaneous 
features of child thought realism in general. For the 
child is a realist, and this means that in almost every 
domain he tends to consider as external, to " reify *' as 
Sully put it, the contents of his mind. And he has a 
systematic propensity for the reification of the contents of 
consciousness that are shared by all minds, whence his 
tendency to materialize and project into the universe the 
realities of social life. 

Without going as far back as Baldwin's " projective " 
stage which is defined precisely by complete realism or the 
indissociation between what is subjective and what is 

1 ClaparMef Experimental Education and Child Psychology. 


objective, we could cite in support of our contention a 
large number of phenomena contemporary with moral 
realism Itself. 

It is particularly necessary to remember at this point 
the definite attitude taken up by children with regard to 
the products or the instruments of thought (see C.W., 
Sect. I.). Dreams, for example, even when the child 
already knows that they are deceptive as to their contents, 
are, till about 7-8, systematically considered as an ob- 
jective reality, as a sort of ethereal, rarefied picture 
floating in the air and fixed before our eyes. Names 
(comparable to moral rules in that they are transmitted 
and imposed by the adult surrounding) constitute an 
aspect of the objects themselves : each object has a name, 
co-substantial with its own nature, having always existed 
and been localized in the object. Finally, thought itself, 
instead of consisting in an internal activity, is conceived 
as a sort of material power in direct communication with 
,,the external universe. 

In the domain of drawing, M. Luquet has given an 
admirable analysis of the phenomenon known as " intel- 
lectual realism *\ The child draws things as he knows 
them to be, not as he sees them. Of course such a habit 
is primarily a proof of the existence and extent of that 
rationalism that belongs to aU thought and which alone 
can adequately account for the nature of perception. To 
perceive is to construct intellectually, and if the child 
draws things as he conceives them, it is certainly because 
he cannot perceive them without conceiving them. But 
to give up gradually the spurious absolutes situated away 
and apart from the context of relations that has been 
built up during experience itself is the work of a superior 
kind of rationality. When the child comes to draw things 
as he sees them, it will be precisely because he has given 
up taking isolated objects in and for themselves and 
has begun to construct real systems of relations which 
take account of the true perspective in which things are 
connected. Thus " intellectual realism ", though it is 


the forerunner of authentic rationalism, also imples a devia- 
tion which consists in isolating too soon and therefore in 
" reifying " the early products of rational construction. 
It is therefore still " realism " in our sense of the term, 
that is to say, it is an illegitimate exteriorization of 
intellectual processes, an illegitimate fixation of each 
moment of the constructive movement. 

Being therefore a realist in every domain, it is not 
surprising that the child should from the first " realize " 
and even " reify " the moral laws which he obeys. It is 
forbidden to lie, to steal, to spoil things, etc. all, so many 
laws which will be conceived as existing in themselves, 
independently of the mind, and in consequence inde- 
pendently of individual circumstances and of intentions. 
For this is the place to recall the fundamental fact that, 
just because of the general realism of his spontaneous 
thought, the child, up to the age of about 7-8, always 
regards the notion of law as simultaneously moral and 
physical. Indeed, we "have tried to show (C.W. and C.C.) 
that until the age of 7-8 there does not exist for the child 
a single purely mechanical law of nature. If clouds move 
swiftly when the wind is blowing, this is not only because 
of a necessary connection between the movement of the 
wind and that of the clouds ; it is also and primarily 
because the clouds " must " hurry along to bring us rain, 
or night, etc. If the moon shines only by night and the 
sun only by day, it is not merely because of the material 
arrangements ensuring this regularity ; it is primarily 
because the sun " is not allowed " to walk about at night, 
because the heavenly bodies are not masters of their 
destiny but are subject like all living beings to rules 
binding upon their wills. If boats remain afloat on the 
water while stones sink to the bottom, this does not 
happen merely for reasons relating to their weight ; it is 
because things have to be so in virtue of the World- 
Order. In short, the universe is permeated with moral 
rales ; physical regularity is not dissociated from moral 
obligation and social rule. Not that the last two are to 


be deemed more important than the first. Far from it. 
There is simply non-differentiation between the two Ideas, 
The idea of physical regularity is as primitive as that of 
psychical or moral regularity, but neither is conceived 
independently of the other. It is only natural, therefore, 
that the moral rule should retain something physical 
about it. Like names, it is a part of things, a character- 
istic feature, and even a necessary condition of the 
universe. What, then, do intentions matter? The 
problem of responsibility is simply to know whether a law 
has been respected or violated. Just as if we trip, inde- 
pendently of any carelessness, we fall on to the ground in 
virtue of the law of gravity, so tampering with the truth, 
even unwittingly, will be caled a lie and incur punishment. 
If the fault remains unnoticed, things themselves will take 
charge of punishing us (see following Chapter, 3). 

In short, moral realism seems to us from this point of 
view to be a natural and spontaneous product of child 
thought. For it is not nearly so natural as one would 
think for primitive thought to take intentions into 
account. The child is far more interested in the result 
than in the motivation of Ms own actions. It is coopera- 
tion which leads to the primacy of intentionality, by 
forcing the individual to be constantly occupied with the 
point of view of other people so as to compare it with Ms 
own. Indeed, one is struck to see how unconscious of it- 
self and how little inclined to introspection is the egocentric 
thought of very young children (J.I?., Chap. IV., i 
and 2). It may be objected to this that primitive thought 
seems, on the contrary, to be directed to a sort of universal 
intentionalism : cMldish animism consists in attributing 
intentions to all things, so also do the " whys ", arti- 
ficialism leads to the notion that nothing exists without a 
motive, etc. But this does not in any way contradict our 
thesis. For to attribute stereotyped intentions to every 
event is one thing, and to subordinate actions to the 
intentions that inspired them is another. The intention- 
alism that characterizes animism, artificialism, and the 


" whys " before 6-7 comes from a confusion between the 
psychical and the physical, whereas the priority of in- 
tentions over external rules implies an increasingly delicate 
differentiation between what is spiritual and what is 

But these considerations are not sufficient to account 
for the phenomena we have observed, and we must now 
turn our attention to the second aspect of moral realism. 
For moral realism is also the product of adult constraint. 
Nor is there, as we have already pointed out, anything 
mysterious in this double origin. The adult is part of the 
child's universe, and the conduct and commands of the 
adult thus constitute the most important element in this 
World-Order which is the source of childish realism. 

But there is more to it than this. It looks as though, 
in many ways, the adult did everything in his power to 
encourage the child to persevere in its specific tendencies, 
and to do so precisely in so far as these tendencies stand 
in the way of social development. Whereas, given suffi- 
cient liberty of action, the child will spontaneously emerge 
from his egocentrisin and tend with his whole being 
towards cooperation, the adult most of the time acts in 
such a way as to strengthen egocentrism in its double 
aspect, intellectual and moral. Two things must be 
distinguished here, differing considerably in theoretical 
importance but of equal moment practically the exter- 
nality of adult commands and the lack of psychological 
insight in the average adult. 

In the first place, moral commands almost inevitably 
remain external to the child at any rate during the first 
years. Most parents burden their children with a number 
of duties of which the reason must long remain incompre- 
hensible, such as not to tell lies of any kind, etc. Even in 
the most modern education, the -child is forced to adopt a 
whole set of habits relative to food and cleanliness of 
which he cannot immediately grasp the why and the 
wherefore. All these rules are naturally placed by the 
child on the same plane as actual physical phenomena. 


One must eat after going for a go to bed at night, 

iiave a bath before to bed, etc., exactly as the 

shines by day and the moon by night, or as pebbles sink 
while boats remain afloat. All these things are and must 
be so ; they are as the World-Order decrees that they 
should be, and there must be a reason for it all. But 
none of it is felt from within as an impulse of sympathy or 
of pity is felt. So that from the first we have a morality 
of external rules and a morality of reciprocity or rather 
of the elements which mill later on be utilized by moral 
reciprocity and so long as these two moralities do not 
unite, the first will almost inevitably lead to a certain 
amount of realism. 

But in the second place, and this unfortunately is no 
less important a consideration, the majority of parents are 
poor psychologists and give their children the most 
questionable of moral trainings. It is perhaps in this 
domain that one realizes most keenly how immoral it can be 
to believe too much in morality, and how much more pre- 
cious is a little humanity than al the rules in the world. 
Thus the adult leads the child to the notion of objective 
responsibility, and consolidates in consequence a tendency 
that is already natural to the spontaneous mentality of 
little children. 

It would be difficult, to be sure, to embark upon an 
objective enquiry in such matters. But if systematic 
investigation is lacking we have some precious sources of 
information which often enable us to plumb greater 
depths than are ever revealed by a mere accumulation of 
incomplete observations. Literature is at hand, moreover, 
to supplement scientific psychology. Edmund Gosse's auto- 
biographic study, Father and Son, not to mention the many 
novels that revive almost unaltered the memories of child- 
hood, tells us more than many a learned treatise on the 
subject. The individual examination of youthful delinquents 
or of " difficult " children is equally illuminating. Finally, 
it is impossible to psycho-analyse an adolescent or an adult 
without discovering that the subject's spontaneous anam- 


nesia (always so full of interest) is crowded with the most 
definite memories relating to the mistakes which Ms 
parents made in bringing him up. 

But although such methods alone will put exceptionally 
illuminating cases within our reach, it might perhaps be 
possible to set afoot an enquiry into the mentality of the 
" average parent " and to accumulate observations made in 
certain homogeneous and comparable situations, such for 
example as those in trains, especially on Sunday evenings 
after a day's outing. How can one fail to be struck on such 
occasions by the psychological inanity of what goes on : 
the efforts which the parents make to catch their children in 
wrong-doing instead of anticipating catastrophes and pre- 
venting the child by some little artifice or other from taking 
up a line of conduct which his pride is sure to make him 
stick to ; the multiplicity of orders that are given (the 
" average parent " is like an unintelligent government 
that is content to accumulate laws in spite of the contra- 
dictions and the ever-increasing mental confusion which 
this accumulation leads to) ; the pleasure taken in inflict- 
ing punishments ; the pleasure taken in using authority, 
and the sort of sadism which one sees so often in perfectly 
respectable folk, whose motto is that "the child's will 
must be broken ", or that he must be " made to feel a 
stronger will than his ". 

Such a form of education leads to that perpetual state 
of tension which is the appanage of so many families, and 
which the parents responsible for it attribute, needless to 
say, to the inborn wickedness of the child and to original 
sin. But frequent and legitimate in many respects as is 
the child's revolt against such methods, he is nevertheless 
inwardly defeated in the majority of cases. Unable to 
distinguish precisely between what is good in his parents 
and what is open to criticism, incapable, owing to the 
" ambivalence " of his feelings towards them, of criticizing 
his parents objectively, the child ends in moments of 
attachment by inwardly admitting their right to the 
authority they wield over him. Even when grown up, he 


be unable, except in very rare to break 

from the affective acquired in this way, will 

be as stupid with Ms own as Ms were with 


It is clearly by this constraint exercised by one genera- 
tion upon the other that we must seek to explain the rise 
and persistence of moral realism. Moral realism, rooted 
as it is in the whole of the child's spontaneous realism, is 
thus consolidated and stylized in a hundred ways by 
adult constraint. Such a meeting of the products of 
adult pressure with those of child mentality is no accident ; 
it is not the exception but the rule in child psychology. 
And this can be only too easily explained, since It is 
through the age-long action, groping its way down the 
centuries, of the generations one upon the other that the 
essential elements of common morality and pedagogy 
have been formed by a mutual adaptation of the two 
mentalities thus confronted. 

In order to show how natural is this double aspect of 
moral realism, let us compare it to a phenomenon which 
is its exact counterpart from the intellectual point of view 
verbal realism, or verbalism, which results from the 
union between the spontaneous linguistic syncretism of 
the child and the verbal constraint of the adult. 

One of the most striking features of the egocentric 
mentality from the intellectual point of view is syn- 
cretism, that is to say, perception, conception and reason- 
ing by general (" global ") l and unanalysed schemas. This 
phenomenon has been described by Decroly and by 
Claparede in the domain of perception, and it reappears 
in every aspect of child thought explanation, under- 
standing, reasoning, etc. (see L.T., Chap. IX.). We found 
it to be particularly prevalent in the domain of verbal 
understanding. A sentence, a story, a proverb will give 

1 In L.T. (ist Ed.) we translated the word global by " general " as 
the use of " global fl had not yet been incorporated into current psycho- 
logical terminology. It means* of course, the opposite of " analysed/' 


the child the impression that he has completely under- 
stood it as soon as he has succeeded in constructing out of 
it a sort of general inclusive schema, or " global " meaning, 
even when individual words or groups of words are still 
quite incomprehensible to Mm. Such an attitude is 
closely bound up with egocentrism. For it is discussion 
and mutual criticism that urge us to analyse things ; left 
to ourselves we are quickly satisfied with a " global ", and 
consequently, a subjective explanation. Now " global " 
syncretism quite naturally leads the child to verbalism. 
Since every word obtains its meaning as a function of these 
syncretic schemas, words end by acquiring a substance of 
their own independently of reality. What, now, are the 
effects of adult constraint with regard to this verbalism ? 
Does it progressively diminish this product of egocentrism 
or does it consolidate it ? In so far as the adult can 
cooperate with the child, that is to say, can discuss things 
on an equal footing and collaborate with him in finding 
things out, it goes without saying that his influence will 
lead to analysis. But in so far as his words are spoken 
with authority, in so far, especially, as verbal instruction 
outweighs experiment in common, it is obvious that the 
adult will consolidate childish verbalism. Unfortunately 
it is the second alternative that is most often realized in 
the teaching given in schools and even in the home. The 
prestige of the spoken word triumphs over any amount of 
active experiment and free discussion. Schools have been 
held responsible for the verbalism of children. This is not 
quite correct, as verbalism arises out of certain spon- 
taneous tendencies in the child. But the school, instead 
of creating an atmosphere favourable to the diminution of 
these tendencies, does base its teaching upon them and 
consolidate them by making use of them. 

All this will have served to show the parallelism between 
moral and intellectual facts in the domain of realism. 
Moral realism and verbalism are therefore the two clearest 
manifestations of the way in which adult constraint com- 
bines with childish egocentrism. 


GENERAL CONCLUSION. The obtained in the 

course of our study of moral realism confirm those of our 
analysis of the game of marbles. There to exist in 

the child two separate moralities, of which, incidentally* 
the consequences can also be discerned in adult morality. 
These two moralities are due to formative processes which, 
broadly speaking, follow on one another without, however, 
constituting definite stages. It is possible, moreover, to 
note the existence of an intermediate phase. The first of 
these processes is the moral constraint of the adult, a 
constraint which leads to heteronomy and consequently 
to moral realism, The second is cooperation which leads 
to autonomy. Between the two can be discerned a phase 
during which roles and commands are interiorized and 

Moral constraint is characterized "by unilateral respect. 
Now, as M. Bovet has clearly shown, this respect is the 
source of moral obligation and of the sense of duty : every 
command coining from a respected person is the starting- 
point of an obligatory rule. This has been abundantly 
confirmed by our enquiry. The obligation to speak the 
truth, not to steal, etc., are all so many duties which the 
child feels very deeply, although they do not emanate 
from his own mind. They are commands coming 
from the adult and accepted by the child. Originally, 
therefore, this morality of duty is essentially heterono- 
mous. Right is to obey the will of the adult. Wrong is 
to have a will of one's own. There is no room in such an 
ethic for what moralists have called et the good J> in 
contrast to " the right " or pure duty, since the good is a 
more spontaneous ideal and one that attracts rather than 
coerces mind. The relations between parents and children 
are certainly not only those of constraint. There is a 
spontaneous mutual affection, which from the first prompts 
the child to acts of generosity and even of self-sacrifice, to 
very touching demonstrations which are in no way pre- 
scribed. And here no doubt is the starting point for that 
morality of good which we shall see developing alongside 



of the morality of right or duty, and which in some 
persons completely replaces it. The good is a product of 
cooperation. But the relation of moral constraint which 
begets duty can of itself lead to nothing but heteronomy. 
In its extreme forms it leads to moral realism. 

Then comes an intermediate stage, which M. Bovet has 
noted with great subtlety I ; the child no longer merely 
obeys the commands given him by the adult but obeys the 
rule itself, generalized and applied in an original way. We 
have observed this phenomenon in connection with lying. 
At a given moment the child thinks that lies are bad in 
themselves and that even if they were not punished, one 
ought not to lie. Here, undoubtedly, is a manifestation of 
intelligence working on moral rules as on all other data 
by generalizing them and differentiating between them. 
But the autonomy towards which we are moving is still 
only half present : there is always a rule that is imposed 
from outside and does not appear as the necessary product 
of the mind itself. 

How does the child ever attain to autonomy proper ? 
We see the first signs of it when he discovers that 
truthfulness is necessary to the relations of sympathy and 
mutual respect. Reciprocity seems in this connection to 
be the determining factor of autonomy. For moral 
autonomy appears when the mind regards as necessary an 
ideal that is independent of all external pressure. Now, 
apart from our relations to other people, there can be no 
moral necessity. The individual as such knows only 
anomy and not autonomy. Conversely, any relation with 
other persons, in which unilateral respect takes place, 
leads to heteronomy. Autonomy therefore appears only 
with reciprocity, when mutual respect is strong enough to 
make the individual feel from within the desire to treat 
others as he himself would wish to be treated. 

And this is the subject we shall try to analyse in the 
course of the next chapter. 

1 See also Baldwin's Social and Ethical Interpretations in Mental 




OUR study of the rules of a game led us to the conclusion 
that there exist two types of respect, and consequently 
two moralities a morality of constraint or of heteronomy, 
and a morality of cooperation or of autonomy. We 
became familiar in the course of the last chapter with 
certain aspects of the first. The second, which will 
occupy us now, is unfortunately much more difficult to 
study ; for while the first can "be formulated in rules and 
thus lends itself to interrogation, the second must be sought 
chiefly among the more intimate impulses of the mind or in 
social attitudes that do not easily admit of definition in 
conversations held with the children. We have established 
its juridical aspect, so to speak, in studying the social play 
of children between 10 and 12. We must now go further, 
and penetrate into the child's actual consciousness. And 
this is where things begin to be complicated. 

But if the affective aspect of cooperation and reciprocity 
eludes interrogation, there is one notion, probably the most 
rational of all moral notions, which seems to be the direct 
result of cooperation and of which the analysis can be 
attempted without encountering too much difficulty we 
mean the notion of justice. It will therefore be on this 
point that most of our efforts wiE be directed. 

The conclusion which we shall finally reach is that the 
sense of justice, though naturally' capable of being rein- 
forced by the precepts and the practical example of the 
adult, is largely independent of these influences, and 

1 With the collaboration of Mies M. Ramberfc, N. Baechler, and 
A. M. Feldweg, 



requires nothing more for its development than the 
mutual respect and solidarity which holds among children 
themselves. It is often at the expense of the adult and 
not because of Mm that the notions of just and unjust 
find their way into the youthful mind. In contrast to a 
given rule, which from the first has been imposed upon the 
child from outside and which for many years he has failed 
to understand, such as the rule of not telling lies, the rule 
of justice is a sort of immanent condition of social relation- 
ships or a law governing their equilibrium. And as the 
solidarity between children grows we shall find this notion of 
justice gradually emerging in almost complete autonomy. 

It was these considerations which prompted us to 
introduce into this chapter the study of a question which 
is not directly connected with the notion of justice that 
of the solidarity between children and the conflicts into 
which this solidarity enters with adult authority in cases 
of " tale-telling ". This analysis will enable us to deter- 
mine at what age solidarity begins to be efficacious. And 
we shall find that it is precisely after this age that the 
equaKtarian notion of justice begins to assert itself with 
sufficient strength to overcome the authority of the adult. 

Finally, to this study of the notion of justice must be 
joined, needless to say, at least a summary analysis of 
children's ideas about punishments. Distributive justice, 
which is defined by equality, has always been connected 
in the general mind with retributive justice, which is 
defined by due proportion between acts and punishments. 
Although the second aspect of the notion of justice is less 
closely connected with the problem of cooperation it must 
also, of course, be submitted to examination. Indeed, we 
shall deal with it first so as to leave our subsequent 
analysis untrammelled by considerations of this particular 
aspect of the question. 

The plan we shall follow is then as follows. We shall 
begin by studying the problem of punishments, then that 
of collective responsibility and of so-called " immanent " 
justice (in which the punishment is supposed to emanate 


from things themselves). After which, by way of transition, 
we shall examine the conflict between retributive justice 
and distributive Justice* Having reached point, we 

shall proceed to the analysis of the relations between 
distributive Justice and authority between childish 

solidarity and authority), then to the study of justice 
between children, and we shall wind up with a general 
discussion of the relations between justice and cooperation. 


JUSTICE. There are in existence two distinct ideas of 
justice. We say that an award is unjust when it penalizes 
the innocent, rewards the guilty, or when, in general, it 
fails to be meted out in exact proportion to the merit or 
guilt in question. On the other hand, we say that a 
division is unjust when it favours some at the expense of 
others. In this second acceptation of the term, the idea 
of justice implies only the idea of equality. In the first 
acceptation of the term, the notion of justice is inseparable 
from that of reward and punishment, and is defined by 
the correlation between acts and their retribution. 

It seems to us more profitable to begin with the first of 
these two ways of thinking because it is the one we can 
relate most directly to adult constraint and to the problems 
examined in the last chapter. It is also very probably the 
more primitive of the two conceptions of justice, if by 
primitive is meant, not so much what is early in point of 
time but what is most overlaid with elements that will be 
eliminated in the course of mental development. For there 
exists in certain notions about retribution a factor of 
transcendence and obedience which a more autonomous 
form of morality tends to eliminate. In any case, the 
problem is to determine whether these two notions develop 
pari passu or whether the second does not tend to pre- 
dominate over the first. 

But in any interrogation of children on punishments we 
are, of course, up against considerable technical difficulties, 
because on such a subject the child is far more likely to 


offer the questioner the usual little family lecture than to 
show him Ms real feelings on the question feelings which 
he but rarely has occasion to put into words, and which, 
perhaps, do not altogether admit of formulation. We 
therefore adopted a more oblique method of approach. 

In order to find out up to what point children regarded 
punishments as just, we decided to split up the diffi- 
culty. In the first place, one can, without throwing any 
doubt upon the notion of retribution itself, suggest dif- 
ferent types of punishment to the child and then ask him 
which is the most fair. In this way it is possible to 
contrast with expiatory punishment the only true punish- 
ment in the eyes of those who believe in the primacy of 
retributive justice 1 a punishment by reciprocity which is 
simply derived from the idea of equality. It goes without 
saying that the child's reaction to such problems as these 
will be highly instructive from the point of view of the 
evolution of the retributive idea. In the second place, 
and once this point has been gained, it will be possible to 
try and find out whether the child regards punishment as 
just and as efficacious by making him compare (in pairs) 
stories where the children are punished and stories where 
the parents are content to reproach their children and 
explain to them the consequences of their actions. The 
subject is then asked to say which of these children will 
be most likely to commit the forbidden deed again, those 
who have or those who have not been the object of 

Once these points have been cleared up but not till 
then it will be possible, in talking to the child, to enlarge 
a little upon the subject and thus lead him on to general 
questions such as what punishments are for, whether there 
is any reason for retribution, and so on. This discussion, 
which would be purely verbal if it came at the beginning, 
can be maintained on the concrete plane in so far as it 

1 See now Durkheim (Education Morale, pp. 188192) revives and 
rejuvenates tbe doctrine of expiation in support of his doctrine of 


draws upon the judgments which the child has Just 
pronounced concerning the stories he told. 

Very briefly, the result we shall be led to is the follow- 
ing. Two types of reaction are to be found with regard 
to punishment. Some think that punishment is just and 
necessary ; the sterner it is, the juster, and it is effi- 
cacious in the sense that the child who has been duly 
chastized will in the future do his duty better than others. 
Others do not regard expiation as a moral necessity ; 
among possible punishments those only are just that 
entail putting things right, a restoration of the stains 
quo ante, or which make the guilty one endure the 
consequences of his deed ; or again, those which con- 
sist in a purely reciprocal treatment. Indeed, apart from 
such non-expiatory penalties, punishment, as such, is 
regarded as useless, reproach and explanation being 
deemed more profitable than chastizement. On the 
average, this second mode of reaction is found more 
frequently among the older children, while the first is 
oftener to be found among the little ones. But the first, 
favoured as it is by certain types of family life and social 
relationships, survives at all ages and is even to be found 
in many adults. 

Here are the questions we used concerning the various 
types of punishment. We begin by saying to the sub- 
ject, " Are the punishments given to children always very 
fair, or are some fairer than others ? " The child generally 
takes the latter view, but whatever he may answer, we 
proceed : " You know, it isn't at all easy to know how to 
punish children so as to be quite fair. Lots of fathers and 
teachers don't know how to. So I thought I would ask 
the children themselves, you and your friends. I shall tell 
you all sorts of silly things that little children have done, 
and you'll tell me how you think they ought to be pun- 
ished/' Then we tell the first part of the story (the 
account of the misdeed committed). The child invents a 
punishment, which we take note of, and we go on to 
say, " Yes, that might do. But the father didn't think of 


that. He thought of three punishments, and he wondered 
which would be most fair. Ill tell you what they were, 
and you will choose which it is to be. 1 ' Care must be 
taken, once the child has' chosen the best punishment, 
to ask him why it is the fairest. Then we ask in the 
child's own terminology which is the most severe (or 
the " stiffest " or the most " boring ", etc.) and de- 
termine whether the child evaluates the punishment in 
terms of its severity or in accordance with some other 
criterion of retribution. 
Here are the stories : 

Story I. A little boy is playing in his room. His mother 
asks him to go and fetch some bread for dinner because 
there is none left in the house. But instead of going 
immediately the boy says that he can't be bothered, that 
he'll go in a minute, etc. An hour later he still has not 
gone. Finally, dinner time conies, and there is no bread 
on the table. The father is not pleased and he wonders 
which would be the fairest way of punishing the boy. He 
thinks of three punishments. The first would be to forbid 
the boy to go on the roundabouts the next day. There 
happened to be a fair the next day, and the little boy 
was to go and have a good time on the roundabouts. 
Well, as he wouldn't go and fetch the bread, he shan't go 
to the fair. The second punishment the father thought of 
was not to let the boy have any bread to eat. There was 
a little of yesterday's bread left in the cupboard and the 
parents are going to eat that, but as the little boy didn't 
go and fetch a fresh loaf, there will not be enough to go 
round. So the boy gets hardly any dinner to eat. The 
third punishment the father thinks of is to do to the boy 
the same thing as he had done. The father would say to 
him, " You wouldn't help your mother. Well I am not 
going to punish you, but the next time you ask me to do 
anything for you, I shall not do it, and you will see how 
annoying it is when people do not help each other." The 
little boy thinks that will be aU right, but a few days later 
he wants a toy that is right at the top of the cupboard. 
He tries to reach it, but he is too small ; he gets a chair, 
but it is still too high. So he finds his father and asks for 
his help. Then the father says, " Now, you remember 
what I said to you, old man. You wouldn't help your 


mother, so 1 don't feel inclined to you either. When 
you make yourself useful 1 shall too, but not before." 
Which of three punishments was the fairest ? 

Story II. A boy has not his sums for school. The 

next day he tels the teacher that lie couldn't do his sums 
because he was ill. But as he rosy cheeks the 

teacher thought that he was making it up, so she went 
and told Ms father and mother. The father wants to 
punish the little boy, but he can't decide between three 
punishments. First punishment : to copy out a poem 
fifty times. Second punishment : the father will say to 
the boy, " You say you are ill. Very well then, we shall 
take care of you. You shall go to bed for a whole day and 
take a dose of medicine to make you better/ 1 Third 
punishment : " You have told a lie. Now I shall not be 
able to believe you any longer, and even if you tell the 
truth I shall not be sure." The next day the boy gets ^a 
good mark at school. Whenever he gets a good mark his 
father gives Mm a penny to put in Ms money box. But 
this time when the boy comes home and says he has had a 
good mark the father says, " That may be true, old man, 
but as you told a lie yesterday I can't believe you any 
longer. I won't give you a penny to-day because I don't 
know whether what you are telling me is the truth. If 
you go several days without telling any lies then I shall 
believe you again and everything will be all right." 
Which is the fairest of these three punishments ? 

Story III. One afternoon a little boy was playing in Ms 
room. His father had only asked Mm not to play ball for 
fear of breaking the windows. His father had hardly gone 
when the boy got Ms ball out of the cupboard and began 
to play with it. And bang goes the ball against a window 
pane and smashes it ! When the father comes home and 
sees what has happened he thinks of three punishments : 
i To leave the window unmended for several days^ (and 
then, as it is winter, the boy will not be able to play in Ms 
room). 2 Make the boy pay for having broken the 
window. 3 Not to let him have Ms toys for a whole week. 

Story IV. A boy has broken a toy belonging to Ms little 
brother. What should be done ? Should he i give the 
little fellow one of Ms own toys ? 2 pay for having it 
mended ? 3 not be allowed to play with any of his own 
toys for a whole week ? 

Story V. Playing ball in a passage (it was forbidden) a 
boy knocked over a pot of flowers and broke it. What 


should be Ms punishment ? i To go into the wood and 
get a new plant and plant it himself ? 2 To be smacked ? 
3 To have all his toys broken on purpose ? 

Story VI. A child Is looking at a picture book belonging 
to his father. Instead of being careful, he makes spots on 
several of the pages. What shall the father do ? i The 
child will not go to the cinema that evening. 2 The 
"father will not lend him the book any more. 3 The child 
often lends Ms stamp-album to his father, the father will 
not take care of it as he has always done up till then. 

Story VII. The leader of a band of robbers has died. 
Two candidates, Charles and Leon, stand for election to 
the leadership. Charles is elected leader. Leon is furious 
and denounces Charles to the police by means of an 
anonymous letter in wMch he accuses him of a robbery in 
which the whole band were involved. He tells the police 
where and how Charles is to be found. Charles is arrested. 
The robbers decide to punish Leon. What should they 
do ? i Not give him any money for a month ? 2 Turn 
Mm out of the band ? 3 Accuse him too of being an 
accomplice by means of an anonymous letter ? 

Obviously the child is not asked all these questions at 
once, but only those that interest him. Equally obvious 
is the fact that these stories are extremely naive, and that 
in real life many of the punishments suggested here 
would be applied according to a very different measure. 
But the essential thing to aim at in these interrogatories 
is to get the stories schematized, even if at times tMs 
means rather forcing the note, and presenting to the child 
types of punishment based on principles that are clearly 
distinct from each other. For it is with the principle that 
the conversation with the subject must concern itself and 
not with the detailed mode of its application. 

Now it appears to us that the punishments described in 
tMs chapter, as likewise punishments in general, can be 
classed according to two distinct principles. Every action 
that is judged to be guilty by a given social group consists 
in the violation of the rules of the group and therefore in 
a sort of breach of the social bond itself. Punishment, as 
Durkheim has shown, therefore consists in restitution, in 
" putting things right ", in a reinstatement of the social 


bond and of the authority of rales. But as we have 
recognized the existence of two types of rules correspond- 
ing with two types of relationship, so 
we must expect to meet in the domain of retributive 
Justice with two modes of reaction and two types of 

There are, in the first place, what we call expiatory 
punishments, which seem to us to go hand in hand with 
constraint and the rales of authority. Take any given 
rule imposed upon the individual's mind from without, 
and suppose the individual to have transgressed this rule. 
Independently even of the indignation and anger that will 
occur in the group or among those in authority, and which 
will inevitably be visited upon the transgressor, the only 
way of putting things right is to bring the individual back 
to his duty by means of a sufficiently powerful method of 
coercion and to bring home Ms guilt to Mm by means of a 
painful punishment. Thus expiatory punishment has an 
arbitrary character, arbitrary in the sense that, in lingu- 
istics, the choice of a sign is arbitrary in relation to the 
thing signified, that is to say, there is no relation between 
the content of the guilty act and the nature of its punish- 
ment. It is all one when a lie has been told whether you 
inflict corporal punishment on the transgressor, or take 
Ms toys away from him or condemn Mm to some school 
task : all that matters is that a due proportion should be 
kept between the suffering inflicted and the gravity of the 

And there are, in the second place, what we shall call 
punishments by reciprocity in so far as they go hand in hand 
with cooperation and rules of equality. Take any rule 
that the child accepts from within, that is to say of wMch 
he knows that it binds Mm to his equals by the bond of 
reciprocity (e.g. not to lie, because lying does away with 
mutual trust, etc.). If this rule is violated, there is no 
need, in order to put things right again, for a painful 
coercion wMch will impose respect for the law from 
without ; it will be enough for the breach of the social 


bond incurred by the transgressor to make its effects felt, 
in other words, it win be enough if the principle of reci- 
procity be brought into play. Since the rale is no longer, as 
before, something imposed from without, something which 
the individual could dispense with, but on the contrary 
constitutes a necessary relation between the individual 
and those around Mm, it suffices to make plain the conse- 
quences following upon the violation of this law in order to 
make the individual feel isolated and to make him long for 
a return to normal relations. Censure no longer needs to be 
emphasized by means of painful punishment : it acts with 
full force in so far as the measures taken by way of recipro- 
city make the transgressor realize the significance of his 
misdeeds. 1 In contrast, then, to expiatory punishment, 
punishment by reciprocity is, to use the terminology of 
linguistics again, necessarily " motivated " ; misdeed and 
punishment, that is to say, are related both in content and 
nature, not to speak of the proportion kept between the 
gravity of the one and the rigour of the other. And so it 
follows that, various kinds of misdeeds being possible, we 
can distinguish in punishment by reciprocity a number of 
different varieties which will be more or less suitable and 
just according to the nature of the reprehensible act. 

In order to classify these varieties let us return to our 
stories. To begin with, it is easy to see which are the 
punishments we regard as expiatory : not being allowed 
to go to the fair or the cinema (I and VI), copying a poem 
out fifty times (II), having one's toys taken away (III 
and IV), being smacked (V) or fined (VII). But of course 
any punishment, even some of the others suggested in 
these examples, can take on an expiatory character 
according to the spirit in which it is applied. A child will 
often explain that he has selected a given punishment 

1 It is clear that measures of reciprocity also contain an element of 
suffering. But the pain is not inflicted for its own sake nor is it destined 
to instil respect for the law in the subject's mind. Such suffering as 
does occur (accompanied at times by material disadvantages) is simply 
an inevitable result of the breach of the bond of solidarity. 


because It the most severe, actually Ms 

to be on quite different principles. In such 

the punishment in question Is clearly still of the 

expiatory order. As to by reciprocity, they 

can be classed as follows, going, roughly, from the more 

to the less severe. 

First there is exclusion momentary or permanent 
from the social group itself (story VII). This is the 
punishment that is often resorted to by children amongst 
themselves, as when, for example, they refuse to go on 
playing with an impenitent cheat. It is the punishment 
which one uses in ordinary life when one refuses to play a 
game with a child or take Mm a walk when experience 
has shown that he cannot behave on such occasions. The 
social bond is temporarily broken. 

A second group can be formed by al those punishments 
that appeal only to the immediate and material conse- 
quences of the act : having no bread for dinner when you 
have refused to go and fetch some when there wasn't 
enough in the house (story I), being put to bed when you 
have pretended to be ill (II), having a cold room when 
you have broken the window pane (III). It is of punish- 
ments like this that Rousseau, Spencer and many others 
were thinking when they claimed that the child should be 
educated by natural experience alone. It is true that, as 
Durkheim has shown, the " natural " consequence of a 
misdeed is necessarily a social consequence, viz. the 
censure it provokes. Only Durkheim seems to think that 
censure, in order to be efficacious, should be accompanied 
by an expiatory punishment, when actually the direct and 
material consequences of the deed are often amply suffi- 
cient to fulfil this office. All that matters is that the 
transgressor should realize that this consequence, however 
natural it may be, is approved by the social group. This 
is why we have classed this kind of punishment among 
the punishments by reciprocity. When the child in story I 
has no bread for dinner, and the one in story III is left 
with a room minus a window pane, since the first had 


neglected to get the bread and the second had broken 
the window, what really happens is that the parents 
themselves refuse to put things right ; they respond to 
their children's negligence by refusing to come to their aid. 
The punishment is still, therefore, one of reciprocity, 
Similarly, when the father in story II pretends to believe 
his son when he is lying and sends him to bed since he 
says he is ill, or again refuses henceforth to believe him 
even when he is speaking the truth (punishment III), he is 
really acting on reciprocity. True, the punishment is a 
** natural consequence " of the act, since the consequence 
of lying is either that the liar is taken at his word, or that 
he is no longer believed at all, but added to this is the 
fact that the father deliberately simulates credulity or 
suspicion in order to show the child that the bond of 
mutual trust is broken. So here again, the principle of 
reciprocity is at work. It is our belief, therefore, that all 
cases of so-called natural punishments imply reciprocity, 
because in all of them the group or the educator wishes 
to make the transgressor feel that the bond of solidarity 
is broken. 

In the third place, there is the kind of punishment that 
consists in depriving the transgressor of the thing he has 
misused. For example, not to lend a child a book that he 
has made spots on (story VI). We' have here a mixture 
of elements analogous to those which characterized the last 
two varieties : a sort of termination of contract owing to 
the conditions ,of the contract not having been observed. 

Fourthly, we can group under the name of simple 
reciprocity, or reciprocity proper, those punishments that 
consist in doing to the child exactly what he has done 
himself. For example, not to help him (story I), to break 
his toys (story V), not to take care of his stamp album 
(story VI), to tell on him, if he has told on you (story 
VII). It need hardly be pointed out that this kind of 
punishment, while it is perfectly legitimate when we want 
"to make the child understand the results of his actions, 
becomes irritating and absurd when it only means giving 


back evil for evil, and capping one Irreparable destruction 
with another (stories I and VI). 

Fifthly, there is purely " restitutive " punishment 
paying for or replacing a broken or object, etc. 

Durkheim was right in contrasting restitutive and retribu- 
tive punishments. But if we divide the latter themselves 
into two types according as they are expiatory or due 
merely to reciprocity, restitutive punishment may be 
taken as the limiting case of punishment by reciprocity. 
For here ceaisure no longer plays any part and justice is 
satisfied with a simple putting right of the material 
damage. Restitutive punishments, it should however, be 
noted, are often impure, and may retain an element of 
retribution. This^is why we have classed them in the 
present division* 

Finally, a sixth category might be distinguished which 
would consist of censure only, without punishment, not 
imposed by authority, but concerned only to make the 
transgressor realize how he has broken the bond of soli- 
darity. But we must not complicate matters unneces- 
sarily, and the question will be reserved for discussion 
later on. 

The conclusion of this analysis is therefore that there 
are two types of punishment or retributive justice: 
-expiatory punishment inherent in the relations of con- 
straint, and punishment by reciprocity. Let us turn to 
the experimental data, and see whether the child tends 
towards one or other of these according to the level of his 

Mile Baechler undertook to question 65 children between 
6 and 12 years of age on these points, and I interviewed 
some thirty myself. The following statistics therefore 
deal with about a hundred children. But as each child 
may give answers that differ according to the story and 
will thus oscillate between expiatory punishment and 
punishment by reciprocity, we calculated by stories and 
not by children, taking as our unit each separate answer 
given by each separate child. As children can hardly be 


questioned on more than four stories at a time, this gives 
us a total of about 400 units. 

It is obvious that in such a field as this, evolution does 
not proceed in exact step with age : there are too many 
interfering factors at work. But taking things broadly, 
we were surprised at the clearness which marked the 
evolution. We divided the children into three groups, 
aged respectively 6-7, 8-10, and 11-12 years (the last 
group including two backward subjects of 13), and we 
found the percentage of punishments by reciprocity out of 
the totality of answers to be as follows : 

6-7 yrs. 8~io yrs- 11-12715. 

Children interviewed by Mile B. 30% 44% 78% 
TOTAL. . . . 28% 49% 82% 

These figures, it need hardly be pointed out, cannot be 
regarded as of great importance. To begin with, they relate 
only to the children belonging to a certain ethnical group 
and a certain social stratum (the poorer parts of Geneva 
and a few children from an elementary school at NeucMtel) . 
Again, it cannot be denied that, however careful one may 
be (see C.W., Introduction), the manner of asking the 
questions plays a very important part. Indeed, it is 
rather disturbing to find that the children one interviews 
oneself answer more often in conformity with one's own 
theory than do the children interviewed by other people ! 
The personal factor here cannot be disregarded, and it 
renders all such statistics more or less open to suspicion. 
All we shall take from these figures is the fact that, 
roughly speaking, judgments of retributive justice seem to 
undergo a certain evolution as the child grows older : the 
younger children favour expiatory punishment, while the 
older ones tend more towards punishment by reciprocity. 

Two important reservations must, however, be made. 
The first is that alongside of the problem of stages, there 
is the problem of types. Some minds are irrevocably 
attached to the idea of expiation (Joseph de Maistre as 
compared to Guyau . . .). They are, of course, the 


product of a certain of upbringing, of a certain Mud 

of social and reHgions education. their deepest 

features subsist independently of age. We would therefore 
not be at all surprised if in the 

of the interrogatory turned out to be quite different. 

In the second place, when the children have to think of 
a punishment themselves, instead of choosing from among 
the various punishments proposed, they nearly always 
turn to expiatory punishment, and their choice is often 
astonishingly severe l . But this does not really contradict 
our present results. Of course, if the child's attention is 
not drawn to the different possible types of punishment 
without even defining them, simply by presenting them to 
Mm as we do here the subject will do no more than think 
of the punishments to which he is accustomed, namely those 
that are " arbitrary " and expiatory. 

Let us now turn to the analysis of the cases. Here, to 
begin with, are examples of children who think arbitrary 
punishments are the most " fair ". 

ANG (6) repeats story I correctly : " How should he be 
punished ? Shut him up in a room. What will that do 
to him tHe*d cry. Would that be fair ? Yes." He 
is then told of the three possible punishments : " Which 
is the fairest ? I'd have not given him his toy. Why ? 
He'd been naughty. Is that the -best of the three punish- 
ments ? Yes. Why ? Because he was very fond of his 
toy. Is that the fairest ? Yes.' 7 Thus it is not the 
principle of reciprocity that carries the day, it is the idea 
of the severest punishment. 

FIL (6), Story I : " My daddy is bad to us. He puts us 
in a dark cupboard, he does. Well, if I had a dark cupboard 
I'd put him in there till the evening and I'd give him a damn 
good box on the ear. And if I had a strap, I'd do nothing 
but beat him." He chooses the third of the three punish- 
ments " because he'd like to go to the fair, -so it'll annoy 
him " (Fr. fa I'&nerve alors). 

ZIM (6), Story I : Zim does not think much of the last 
two punishments. The third " is not hard. Why ? 
On the little boy. Why is it not hard on him ? It isn't 

1 See, in this connection, the enquiries made by Knapp in V Inter- 
m&diaire des Educaieurs. 



much." The second is also " not much ". The fairest is 
therefore the first " because Ms not on the Roundabouts ". 

MORD (7), Story VI : The fairest punishment is to stop 
him going to the cinema. " Why ? It suits Mm better 
than the other two. It's worse. It's lovely going to the 
Pictures. But it's not lovely not giving him the album back. 
Why ? Going to the Pictures is lovely. But when he's 
had his book five or six times he'tt say, l I've had enough of 
looking at it, I don't care if he doesn't knd it to me any 
more *." 

SYL (7|), Story II : " The fairest ? To make her do 
fifty lines. That's the worst punishment, because she isn't 
allowed out'' 

MAY (7!), Story I : The fairest punishment is " not to 

five him any bread. That'll punish him most. That'll make 
im go and fetch some other time. 39 Story VI : "I wouldn't 
take her to the Pictures, because that is what she would like 
lest, the Pictures. What is the second punishment ? I 
wouldn't take care of her album. She likes collecting stamps. 
Would that do ? It wouldn't punish her enough. It 
wouldn't make her [become] good enough. Which is the 
worst punishment ? Not taking her to the Pictures!' 

Au (7|), Story II : " I'd make him write out fifty lines 
in his copy book. That would be a punishment, then he 
wouldn't do it again because he'd have to write them out 
again. l s that the fairest ? It serves the little boy right. 
He didn't need to tell lies. [It is the fairest] because it is a 
strict punishment. Which is the fakest? Writing out 
fifty lines because it's a bore. He can't have any fun." 

BLA (7!), Story I : " 7 wouldn't have let him go on the 
Roundabouts. Why ? Because it's lovely on the Rounda- 
bouts ! " 

PEL (7i), G., Story I: "Which do you think the 
fairest ? Not to go on the Roundabouts. Why ? Because 
he didn't help his mother. Which is the ' stiff est ' of the 
three? Not to go on the Roundabouts/ 9 Story II: ''Which 
is the fairest of the three punishments ? Copying out the 
poem fifty times. Why is that the fairest ? Because it is 
the strictest" 

JEAN (8), Story I : " Which of the three punishments 
was the fairest ? Not to go on the Roundabouts. Why was 
that the fairest ? Because the child wants to go [on the 
Roundabouts] and he's not allowed to. Which of the three 
punishments does he hate most ? Not going on the 


Sur (8), same reaction to the first story. Story II : 

" Which Is the fairest of those three ? 

When he to copy out ike times. Why 

was that the fairest ? he to his 

sums, he didn't. Which Is the worst bore of the 

three ? Copying out the fifty times Story IV : 

" Which of these ponishments do you think is the fairest ? 
The third [taking Ms toys away]. Why Is that the 
fairest ? he shouldn't Ms brother's 

toy. And are the other two fair too ? Yes, M'sicu. Let 
us take the second and third. Which is fairest, to make 
him pay for the toy he broke or to take all his toys away 
from Mm ? To take all his toys away. Why Is that 
fairer ? . , . What does he hate most ? Having his 
toys taken away." Story V : " Which Is the fairest ? 
Having a toy broken. Why is that the fairest ? . . . 
Which would have made him most angry of these three 
punishments ? To have one of his toys broken. 93 Thus 
what counts Is not reciprocity, even In this last case ; It is 
the idea that the severity of a punishment determines its 

KEC (8), Story I : " Which is the fairest ? Not to go on 
the Roundabouts. Is that the fairest ? Yes. Why ? 
Because he likes going on the Roundabouts." As to the 
other two, the fairest would be not to let him have any 
bread. " If he likes bread, then he mustn*t be given any" 

BAD (9), Story I : " I like the one about the Cinema 'best. 
Because that is the fairest, because he will have been stopped 
doing something he likes.' 1 

BAU (10), Story I : " The Roundabouts is the best. 
Why ? Because he liked that. . He ought to be stopped doing 
what he liked doing best." 

The general meaning of these answers is not hard to 
perceive. In these children's eyes, punishment consists, 
as a matter of course, in inflicting upon the guilty a pain 
that will smart enough to make them realize the gravity 
of their misdeed. Naturally, the fairest punishment will 
be the most severe. Each of the subjects questioned 
marks in his own way this linkage of the idea of 
retributive justice to the severity of the punishment, but 
the most characteristic expressions are, " It is worse *' 
(Mord, Syl, etc.), " That will punish him most " (May), 
" A strict punishment " (Ali). 


It is quite clear that none of these children mean the 
punishment to mark a break in the bond of solidarity nor 
to drive home the need for reciprocity : there is a pre- 
dominance of expiatory punishment. True, there exists 
a certain ambiguity on this point. Many educational- 
ists look upon punishment, even when it consists in 
inflicting " arbitrary " suffering, only as a preventive 
measure, aimed at avoiding repetitions of the fault. Only 
a minority regard punishment as strictly expiatory, i.e. 
as serving to wipe out either by compensation or by the 
efficacy of suffering the actual fault that had been com- 
mitted. How does the matter stand with children ? The 
remarks made on punishment in general by the subjects 
questioned in the preceding interrogatory seem to confirm 
us in believing that the two attitudes coexist in each child, 
but in a confused and undifferentiated manner. For the 
child will at one time emphasize the vindictive aspect of 
punishment as of sheer chastisement inflicted by a higher 
power (see Fil, for example), at others he comes of himself 
to the theory of preventive punishment. Thus, according 
to May, a given punishment is not sufficient because " it 
wouldn't make her [become] good enough ". But even 
here there is in the child's mind the idea of a necessary 
compensation, and it would be contrary to justice in his 
view not to punish the transgressor at all. And as the 
punishments in question are chosen as a function of the 
painful element they contain, this necessary compensation 
is thus equivalent to the notion of expiation. 

Let us now turn to the children who regard punishment 
by reciprocity as more just : 

GEO (7), Story I : " Which of the three punishments is 
the fairest ? Not to help Aim. Why IHe hadn't helped 
at home, so it is almost the same thing. And if his father 
hadn't thought of that punishment, which would be the 
fairest ? Not to go on the Roundabouts . . . Oh, no ! 
That he should have no supper. Because he wouldn't help 
his mother, so he mustn't have any supper. -Which of the 
three punishments is the least fair ? The one of the 


Why ? A0 to 

Story III : " Which Is the ? To ^zy >r 

& pane. Why Is it the fairest ? it 

be as if he for it [ = it would be 

patting right'.' And otherwise, would be the 

fairest ? To the Thd 

him not to break Which is the 

least fair ? To take his away for days. Which 

is the worst bore ? Not to play" 

DESAR (7$), Story I : " Which of punishments is 

the fakest ? That any bread. 

Why is that the fairest ? Because he didn't go to fetch any. 
Which is the * stiffest ' of these three punishments, the 
one he would like to have least ? Not to go to the Round- 
abouts. Then, which is the fairest ? Not to give him any 
bread. Why do you think that the fairest ? Because he 
didn't go awl fetch the bread." Desar is fully aware of the 
relation of cause to effect that enters into this punishment, 
but he does not succeed in making it explicit. Story II : 
The fairest would be not to believe him any more. 
" Why ? Because not to believe him any longer would be 
quite true, because he told his school teacher a lie." Story 
III : " Which do you think would have been fairest ? 
To pay for the window pane. Why ? Because it wouldn't 
be fair for the mother and father to pay. What was the 
punishment that the little boy hated most, to pay for the 
window pane, or to be in a cold room ? To have a broken 
pane [to be cold]/ 1 Story V : " Which is the fairest ? 
To break one of his toys. Why is that the fairest ? To 
break one of his toys because he broke the pot. Which of the 
three would he have hated most ? To go and get a plant 
in the forest. And which was the fairest ? To have one of 
his toys broken.' 9 Desar, it will be seen, leans towards 
reciprocity throughout, even in the paradoxical case of 
story V. 

BERG (8), Story I : "He ought not to have been sent to 
the Roundabouts. Would that be the fakest punishment ? 
No, that's not fair. It ought to have been the toy he was 
stopped from [reaching in the cupboard that should not have 
been given to him], because he didn't help he oughtn't to be 
helped either. Is that the best punishment? Yes. He 
didn't help so they oughtn't to give him any help either. 

BAUM (9), Story I : " The last is the best. Since the boy 
won't help, well his mother won't help him either. And 
which is the fairest of the other two punishments ? Not 


to give him any bread, then hed have nothing to eat at supper, 
because he wouldn't help his mother. And the first ? Thai 
was the one he deserved least. He wouldn't have minded. 
He'd still have been able to play with his toys [ = lie would 
still have been helped by having Ms toy fetched for Mm] 
and he would have had bread in the evening." Story VI : 
" / would dirty his album for him, because that would be the 
fairest punishment. It would be doing the same to him as he 
did f And of the other two, wMch is the fairest ? I 
wouldn't have lent him the book again because he would have 
made spots on it again. And how about the first punish- 
ment * to stop him going to the Cinema ' ? That one is 
the least fair. It does nothing to the album. It doesn't 
change the album, the book. It has nothing to do with the 

DEC (gj), Story I : The fairest is the one of the toy. 
" Why ? He didn't help his mother, so why should his 
mother help him.'* 

RID (10), Story I : The best is " the one of the toy, 
because it is to show him how one likes it when people 
don't give help. Which is the fairest ? The one of the toy, 
because his mother does the same thing to him as he had done 
to her." Story II : " WMch is the fairest ? The one when 
he was ill [ = when he was put to bed], because as he said 
so [that he was ill] you had to believe it. And of the two 
others, which is the fairest ? The one about the twopence, 
because since he liked having his twopence and had told a lie, 
then you oughtn't to believe him any longer. And the first 
(copying fifty times) ? That's a bit thick. . . . Which of 
the three is the fairest ? There are two more or less fair. 
The one of pretending fye is ill and the one of not believing 
him" Story VI : " Which is the fairest INot the one of 
the Cinema, because that's rather too strict for having made 
spots. And wMch of the other two ? The one of making 
spots on his album . . . it was right to do to him what he 
had done. 3 ' 

Nus (n), Story I : " I'd have given him a smacking." 
The father thought of three punishments. (I tell them to 
him.) " Which do you think is the fairest ? Not to give 
him any help. Do you think that is fairer than smacking 
4 him? Fairer. Why? (He hesitates) . . . Because it's 
doing about the same thing to him as he had done. And of 
the other two, wMch is the fairest ? Not to let him have 
any bread. Why ? Because he didn't fetch any." 

ROY (n), Story I : " WMch punishment do you think 


the ? Xoi 

Why Is it ? If s the thing. 11 11 : 

The fairest is " not to the he 

He a lie them.*" 

Story VI : " Not to cf Ms the boy 

care" Story VII : " I'd have a 

he a- too." 

BUH (12 J), Story I : " punishment is the 

fairest ? The one of not his supper. 

Why ? Because he go gel Which is the 

punishment he will hate the most ? Not going on the 
Roundabouts. And the fairest ? Not any bread, 

Is not to go to the Roundabouts as f ak as the other pun- 
ishment or less fair ? Less fair. Why ? You 
give him help. [The Roundabouts punishment is less 
fair] because there is no ih& bread and 

Roundabouts." Story II : " WMch do you think is the 
fairest ? To make him go to bed. Why ? Because he 
tried to make them believe he ill. And which of the 
other two is the fairest ? Not to believe him any longer. 
Why ? Because he told a lie. WMch is the punishment 
that has no connection ? Copying out a problem Jifty 
times. And which has most connection ? Putting him to 
bed. And what about a fourth which would be not to 
punish Mm at al, would that do ? He ought to be punished 
att the same! 9 Story IV : The fairest punishment is " that 
he should give one of his toys to the little boy. Did you 
choose that one just because it came into your head, or 
because it seems to you more just ? He took a toy away 
from the little boy, so it is right that he should give one back 
to Mm." 

It will be seen how different these cMldren's reactions 
are from those of the preceding group. The value of a 
punishment is no longer measured by its severity. The 
essential point is to do to the transgressor something 
analogous to what he has done himself, so that he should 
realize the results of Ms actions ; or again to punish, 
where it is possible, by the direct material consequences 
of Ms misdeed. Sheer reciprocity stands so Mgh in the 
eyes of the child that he wiU apply it even where to us it 
seems to border on crude vengeance, as in breaking a toy 
(Story V), etc. The reason for this is, as we shall see later 


on, that between 7 and 10 sheer equality in al its bru- 
tality still outweighs equity. 

A problem of interpretation arises nevertheless : have 
the answers quoted above really any moral significance, or 
was the child's intelligence alone concerned ? For the 
supposition might be made that the child regards the 
question put to him simply as a test of his intelligence 
and seeks among the suggested punishments those that 
have some connection with the act, precisely because he 
has been asked to make a choice. In other words, he 
thinks more or less as follows : " They're suggesting three 
punishments to me, so there's a catch in this somewhere. 
Now some of them are connected with the action and some 
are not. Let's choose the one that is most like the mis- 
deed, and then see whether that is the answer they want/' 
Thus the choice would be dictated by the intelligence 
alone and not by a sense of justice. 

But though the intervention of this factor could not, of 
course, be excluded from the interrogatory, we believe the 
tone of the answers to be primarily moral. When Geo 
(Story I) compares punishments by reciprocity and expi- 
atory punishment, he emphasizes very clearly that the 
first are just, whereas the last is cruel. Note also Dec's 
little argument founded on the principle of sufficient 
reason ! For the rest, not only will the sequel convince us 
of the growing importance of the ideas of reciprocity and 
equality in children between 7 and 12, but educational 
experience is there to tell us how the child reacts in every- 
day life. Now, without wishing to impose one system of 
moral education rather than another we are speaking as 
psychologists and not as educationalists it does seem to 
us to have been demonstrated that those teachers whose 
ideal it is to set cooperation above constraint succeed in 
accomplishing their aim without the use of expiatory 
punishments, and thus prove that children fully grasp 
the meaning of punishment by reciprocity. Even if 
among the very little ones censure and preventive 
measures (taking away from the child an object that 


he Is to break, etc.) are Interpreted 

as expiatory punishments, the more the develops, 

better able he becomes to value of measures 

of reciprocity. And we believe, without to 

the point, that the answers obtained in the course of our 
interrogatory correspond to sentiments that have been 
really experienced by the child, either through Ms having 
In the past proved for himself the soundness of certain 
punishments by reciprocity, or by Ms having felt the 
doubtful character of many expiatory punishments and 
thus being more inclined to approve of the punishments 
by reciprocity proposed in our stories. 

This leads us back to the second point touched upon at the 
beginning of this section, we mean the efficacy of expiatory 
punishments. It Is a striking fact that in the early part 
of the Interrogatory the children are almost unanimous In 
defending severe punishments both as legitimate and as 
educationally useful. They are sincere and eager votaries 
of the current morality. But considering how clearly, 
later on, many of them choose reciprocal as opposed to 
" arbitrary " punishment, may we not go a step further 
and ask whether the child Is really convinced of the utility 
of punishment ? Does he not often feel that a timely 
appeal to his generosity would lead to better results ? 

Let us try to penetrate Into Ms judgment on this point 
with the following experiment. Let us first tell him the 
story of some misdeed chosen from among the usual 
childish faults. Then let us describe the two alternatives : 
on the one hand, a severe expiatory punishment, on the 
other, simple explanation with an appeal made to the 
principle of reciprocity, but not accompanied by any 
punishment whatsoever. And let us then ask the subject 
in which of these two cases a relapse is most likely to 

Here are the stories used to this end : 

Story I, A. " A boy was playing in his room, while his 
daddy was working in town. After a little while the boy 
thought he would like to draw. But he had no paper. 


Then lie remembered that there were some lovely white 
sheets of paper in one of the drawers of his father's desk. 
So he went quite quietly to look for them. He found 
them and took them away. When the father came home 
he found that his desk was untidy and finally discovered 
that someone had stolen Ms paper. He went straight into 
the boy's room, and there he saw the floor covered with 
sheets of paper that were all scribbled over with coloured 
chalk. Then the father was very angry and gave Ms boy 
a good whipping/' 

B. " Now I shall tell you a story that is nearly the 
same, but not quite (the story is repeated shortly, except for 
the last sentence) . Only it ends up differently. The father 
did not" punish him. He just explained to Mm that it 
wasn't right of him. He said, ' When you're not at home, 
when you've gone to school, if I were to go and take your 
toys, you wouldn't like it. So when I'm not there, you 
mustn't go and take my paper either. It is not nice for 
me. It isn't right to do that/ 

" Now a few days later these two boys were each of them 
playing in their garden. The boy who had been punished was 
in his garden, and the one who had not been punished was 
playing in his garden. And then each of them found a 
pencil. It was their fathers' pencil. Then each of them 
remembered that his father had said that he had lost his 
pencil in the street and that it was a pity because he 
wouldn't be able to find it again. So then they thought 
that if they were to steal the pencil, no one would ever 
know, and there would be no punishment. 

" Well now, one of the boys kept the pencil for himself, 
and the other took it back to Ms father. Guess which one 
took it back the one who had been well punished for 
having taken the paper or the one who was only talked 

Story II. A. " Once there was a little boy who was 
playing in the kitchen while his mother was out. He 
broke a cup. When Ms mother came home, he said, ' It 
wasn't me, it was the cat. It jumped up there/ The 
mother saw quite well that this was a lie. She was very 
angry and punished the boy. How did she punish him ? " 
(You leave it to the child to decide upon the punishment.) 
B. Idem. tl But this time the mother didn't 
punish him. She just explained that it wasn't very nice to 
tell lies. ' You wouldn't like it if I were to tell you lies. 
Suppose you were to ask me for some of the cake that's in 



" A few days later, the two little boys 
in their kitchen. And they are with the 

matches. When their mother in, one of 

a lie again, says he was not with the matches. 

The other one owns up at once. Which one it who 
told the lie again, the one who for 

breaking the cup, or the one who only talked 


The stories are, of course, extremely naive. But they 
suffice, in our opinion, to bring out the child's mental orien- 
tation. If he really believes in punishments, he will show 
it. If he simply wants to please us rather than give his 
own thoughts, he will also answer in support of punish- 
ments (since in his eyes there is every chance that a 
gentleman who questions schoolboys will believe in punish- 
ment !). If the child replies in favour of simple explana- 
tion, it is, so it seems to us, because there is something in 
Mm that makes him look upon reciprocal generosity as 
superior to any form of punishment. 

Now, out of the thirty children who were questioned on 
this point alone (not counting the supplementary questions 
put to the hundred children spoken of before) those of 7 
and under declared themselves almost unanimously in 
favour of punishment, whereas more than half the cases 
between 8 and 12 answered in the opposite sense. 

Here are examples of the first type : 

QUIN (6). Repeats Story I correctly: "Which one 
brought the pencil ? The one who was punished. Then 
what happened, did he do it again or not ? Not again. 
And the one his father didn't punish ? He stole again. 
If you had been the daddy, when they stole the paper, 
would you have punished them or explained ? Punished. 
Which is fairest? To punish. Which is the nicest 
daddy, the one who punishes or the one who explains ? The 
one who explains. Which one is fairest, the one who etc. ? 
The one who punishes. li you had been the boy, which 

220 OF THE 

would you have thought fairest, to be or to 

have to you ? 

it had to you, would you have done it 

again ? No. And if you punished ? No, I 

Which of the two boys didn't do it 
again ? The one punished. What is the good of 

punishing ? Because you're a bad boy. 93 

KAL (6), Story II : " WMch one told the lie about the 
matches ? The one the mother punished properly, [Kal 
chooses the dark cupboard as a punishment !] he told the 
truth. And the one who had not been punished ? He 
a lie again. Why ? Because he hadn't been punished. 
Why did the other one not tell a lie ? Because he ha$ 
been properly punished. 39 

SCHMEI (7), Story I : " Guess what the boy who had 
been punished by Ms father did. He gam it back, because 
he afraid his father would scold him again. And the 
other one ? He kept it, he knew his father [thought he] 
had lost it md of doors. Which of the two fathers was the 
most fair ? The one who punished him properly. Which 
of the two fathers was the most of a sport ? The one who 
didn't scold, the one who explained. Which of the two boys 
loved Ms father best ? The one when the father was a 
sport, Which boy was nicest to Ms father ? The one who 
gam the pencil back to his father. Was that the one who 
had been punished, or not punished ? Punished." 

Box. (8), Story I : " WMch one gave it back, the one 
who was punished, or the one who was not ? The one who 
was punished. What did he tMnk ? He thought, 1 1 
don't want to be punished again '. And what did the 
other one think ? He thought, ' As I wasn't punished 
before, I won't be punished this time '. Which of the two 
fathers was the most fair ? The one who punished. If 
you had been the father would you have punished him ? 
/ would have. Would you have whipped him ? I would 
have put him to bed. Which of the two fathers was the 
most of a sport ? The one who didn't punish. Which of 
the two boys was the nicest ? The one who was punished. 
Which one was the nicest, the one whose father was fair, 
or the one whose father was a sport ? The one whose 
father was fair. Supposing you had stolen something, 
would you rather they punished you or explained things 
to you ? Punish me. Should one be punished ? Yes. 
Then the more one is punished the better it is. It makes 
you better." 

OF 221 

Here, finally, is an intermediate is 

as a of the 

FAR (8} $ Story I : " Which one it back ? rfo one 

punished.' Why ? he got And 

the other ? He it, he 

Which of the two fathers is the most fair ? The one 
punished. Which one was most of a sport ? The one 

Why is he more of a sport ? Because he 
explained. Which of the two boys was the nicest ? The 
one punished, Which of the two fathers was 

right ? The one who didn't beat him. Which of the two 
fathers would you have given the pencil back to ? To the 
father didn't punish. Why ? Because he the 

nicest. If you had been the father, what would you have 
done ? 7 wouldn't have him, I ex- 

plained. Why ? So that he shouldn't steal again. Which 
of the fathers was the most fair ? The one who punished. 

I've told you a story, now you tell me one, one that 
really happened, when you were punished. Yes. I ran 
into the field. Where ? Into our field, in the grass. They 
kit me. And then ? I didn't do it again. And if they 
hadn't beaten you ? / would have done it again. Should 
people always be punished ? Always when you've been a 
bad boy: 9 

It will be seen how closely all these children cling to the 
traditional view of punishment as morally necessary qua 
expiation and educationally useful to prevent a relapse 
into evil. The last cases quoted, it is true, consider it 
more " sporting " only to explain* without chastising, but 
this is neither just nor wise. Only Far hesitates for a 
moment towards the middle of the interrogatory, but the 
tradition of his fathers is too strong for him, and he 
reverts to the customary morality. 

Here, on the contrary, is a different set of opinions, 
which may be considered characteristic of a second type 
of moral attitude, and, up to a certain point, of a second 
stage in the social development of the child. 

BRIC (8), Story I : " What did they do ? One of them 
gave it back, the other one kept it. Which one gave it back ? 

The one who was not punished. What did he say to 

222 OF 

? TTwrf he to give it back, he 

the other one IThat he io 

#. Why ? Ad raws punished" The beH 

rings. Brie out for a quarter of an hour's break. 

We " What were we doing before break ? 

a Do you remember what it was ? Yes, 

who had Then afterwards they found 

a pencil, one of it back not the other. 

Which one gave it back ? The one who had not 

punished. What did he say to himself ? That he to 

it back because his father would be pleased. And the 

other one ? He kept it. Why ? Because he didn't 

to "his father. Which of the two fathers would you 

like to be ? The one who explains. And which of the two 

children ? The one who was not punished. Why ? 

he'll that he mustn't steal [since ^ it is 

explained to Mm]. And if he is punished, what will he 

do ? Perhaps he'll try again and then not be punished.** 

SCHU (8), Story I : The boy who gives back the pencil 
is the one who has not been punished. " Why did he give 
it back ? Because they explained to him [about the first 

theft]. Why ? Because it's a better way to make him good. 

Which of the two fathers is most of a sport ? The one 

who explained. And which is fairest, explaining or punish- 
ing ? Explaining. Why did the one who was punished 
begin again ? . . . And if they had explained to him, 
would he have begun again ? No. Why ? Because he 
would ham understood. And wouldn't he have understood 
that you mustn't steal if he had been punished ? He 
wouldn't have understood so well. Now listen to me care- 
fully, I am going to change the story round a little. Let 
us say that things have been properly explained to both 
the boys. But one of them was also punished, and the 
other one was only talked to without being punished. 
WMch of the two -gave back the pencil later on ? The one 
who was not punished. Why ? -Because he had under- 
stood things better than the other one. Why did the other 
one do it again ? Because he hadn't understood things quite 
so well, Why not ? Because he was scolded and explained 
to at the same time. Does your father not punish you ? 
He more often explains. Do you think it fair that you 
should be punished ? No, not fair. Why ? Because I 
can understand much better when people explain things to me. 
Tell me about once when you were punished. Once I 
was staying with my Granny. Fm never punished at home. 

OF 223 

l Granny's I What you ? 

I had a How you IThey 

my And your father not your ? 

Hardly ever" Story II : Same answers. "Which 

didn't do it ? lib one ike to. 

the other one, did it did he say to 

himself inside ? will me but he won't do 

ril tell a lie.* Which of the two 

fathers was a sport ? The one explained, And which 
was the most fair ? The one explained." 

CLA (9), Story I : " Which one gave it back ? The one 

his to, And what did the other say 

to himself ? / may as it. Daddy it'on't see," 

" Which of the two fathers was the fairest ? The one 
didn't punish. Which is fairest, to punish or not to 
punish ? Not to punish.* 9 " If you had been the boy, 
what would you have done ? I'd given it back. And 
if you had been punished ? Td given it back all the 
mine [!]." " Which boy was nicest to his father ? The 
one who gave back the pencil. But ordinarily, everyday, 
which one is nicest to his father, the one who is punished 
often, or the one who is not ? The one that you explain 
things to. Why ? Because v^u don't do it again. Which 
is best, to explain and then punish, or to explain and then 
forgive ? To explain and then forgive.** 

It will appear immediately how different is these chil- 
dren's attitude from that shown by the others. Nor is 
this new reaction merely verbal. Of course the wild 
generalizations to which the interrogatory leads are apt to 
give the impression that the subjects are indulging their 
imaginations in the sugary form of morality current in 
some paradise of good children. But side by side with 
these cases, what psychological penetration some of the 
remarks reveal ! When Schii, for example, seeks to show 
that the child who is punished is more likely than the 
other to begin again, he is obviously thinking of those 
all too frequent cases where the accumulation of punish- 
ments makes the offender insensitive and coldly calculat- 
ing. " Daddy will punish me, but he won't do anything 
to me afterwards ! " How often, indeed, one sees children 
stoically bearing their punishment because they have 


decided beforehand to endure it rather than give in to the 
superior will ! And again when the same Sehil compares 
the punishment meted out to Mm at Ms grandmother's 
with the ordinary reactions of his own father, it is hard 
not to recall the comparisons we have all made in our own 
childhood between the understanding attitude of one 
relative and the unpsychological severity of another. 

It is our belief, therefore, that the answers examined 
above correspond, up to a given point , with the experi- 
ences of real life ; and this would mark the existence of a 
certain evolution with increase of age in the judgments 
made by children on the subject of punishments. On the 
other hand, the interrogatory dealing with the more 
general and abstract question as to the use of punishments 
(and whether they are just, etc.) did not yield anytMng 
of very great interest, for at all ages the answers reflect 
the ideas of the child's surroundings rather than his 
personal feeling on the subject. A difference of attitude 
should, however, be noted between the younger and the 
older children with regard to the justification of punish- 
ment. For the little ones, the idea of expiation is neces- 
sarily bound up with the idea of preventing a relapse. 

TRAP (6) : " Should children be punished ? Y^s, 
when they're naughty, you punish children. What does 
* naughty ' mean ?? It means naughty, when children are 
naughty, when you punish them. Is it fak to punish ? 
Yes 3 because if you've done nothing then it isn't fair, but 
when you have done something [it is fair]. Is it a useful 
thing to punish ? What is it for ? Yes, because they'd 
only got to not disobey, because they were naughty. " 

ZIM (6) ; " Is it fair to punish ? Yes, it's all fair. Is 
it useful to punish ? What is it for ? Yes, it's- a good thing 
to punish them when they're silly ; it's always a good thing.*' 

MAIL (6) : " Is punishing fair ? Yes, because it's always 
fair. Is it useful ? Yes, because it's for when you are 
naughty. What does it do [Fr.fait ?] ? It makes [Fr.fait] 
a punishment [== it chastises]/* 

The older ones, on the contrary, concentrate above all 
and almost exclusively on the preventive utility with very 
definite diminution of the idea of expiation. 


(u) : " Is It fair ? Yes, 

A itof 10 a Is it ? 

i>$, i/ >*OM Ae tfo # 

DUP (n) : " Is It fair ? Yes. Is it ? Yes, 

J/0M to &! YOU fj 

aren't bs 

GUI (12) : " Is it fair? Ys, ij you've 

; are not fair ; to be 

in to the fault. Is It ? OA y^s, so s nrt 

fo again," 

Let us now try to formulate our conclusions. Difficult 
as is the interrogatory on such delicate points, and 
deeply tinged with the phraseology of adult morality as 
are the answers, taking things broadly, the results we 
have obtained do nevertheless seein to us to converge. 
They seem to point to the existence of a sort of law 
of evolution in the moral development of the child. It 
would seem that we have to distinguish in the domain 
of justice between two types of reaction, one founded 
on the notion of expiation, the other on that of reciprocity, 
And though representatives of both types are to be foond 
at all ages, it would seem nevertheless that the second 
tended to predominate over the first. 

The choice of punishments is the first thing that brings 
this out. The little ones prefer the most severe so as to 
emphasize the necessity of the punishment itself; the 
other children are more In favour of the measures of 
reciprocity which simply serve to make the transgressor 
feel that the bond of solidarity has been broken and that 
things must be put right again. It is also brought out by 
the reactions of the subjects whom we questioned on the 
subject of relapses. The little ones think that a well- 
punished child cannot repeat its offence because it has 
realized the external and coercive authority of the rule in 
question, whereas many of the older ones hold that a child 
to whom, even without punishment, the consequences of 
Ms actions have been thoroughly explained is less likely 
to begin again than if he had been punished and nothing 


more. The same thing, finally, would seem to be con- 
firmed by the short interrogatory on the utility and 
soundness of punishments in general: the little ones 
introduce an expiatory element into all their* answers, 
whereas the older children are content to justify punish- 
ments by their preventive value. On this point, indeed, 
these older children take up an attitude that definitely 
contradicts that which they had been observed to hold in 
the last interrogatory. This is because here they feel con- 
cerned to defend in their own way the views that are 
generally held by those around them, whereas in the 
stories about relapsing the answers are more personal and 
more spontaneous. 

These two types of attitude, which we believe we have 
been able to dissociate from one another, are naturally, in 
so far as they correspond to real facts, connected with the 
two moralities which we have so far traced in the behaviour 
and judgment of the child. The notion of expiation 
corresponds, of course, to the morality of heteronomy and 
-duty pure and simple. For one whose moral law consists 
solely of rules imposed by the superior will of adults and 
older children, it follows that the disobedience of small 
children will naturally entail the anger of their elders and 
this anger will take the concrete form of some kind of 
" arbitrary " suffering inflicted upon the offender. This 
reaction on the part of the adult is legitimate in the eyes 
of the child in so far as the relation of obedience has been 
broken and in so far as the suffering imposed is in pro- 
portion to the fault that has been committed. In the 
ethics of authority any other punishment is incompre- 
hensible. And since there is no reciprocity between 
commander and commanded, what happens is that even 
if the former inflicts upon the latter only a " motivated " 
punishment (simple reciprocity, consequences of the act, 
etc.) the child will see nothing in it but an expiatory 
chastisement *. Punishment by reciprocity, on the other 

i We have to thank the head mistresses of the Maison des Petits 
for the following clear confirmation of our statements. They tell us 


hand, corresponds with cooperation and an autonomous 
ethic. Indeed it is impossible to see how the relation of 
mutual respect, which is the foundation of all cooperation, 
could possibly give rise to the idea of expiation or render 
it legitimate. It is, on the contrary, easy to see how 
censure (which is at the origin of any punishment whatso- 
ever) can, in the case of cooperation, be accompanied by 
definite measures which have been taken in order to mark 
the break in the bond of reciprocity or to make the 
offender understand the consequences of his acts. 

If, then, we admit this kinship between the two types 
of attitude relating to retributive justice and the two 
moralities which we have distinguished up to the present, 
what explanation are we to give of the genesis and destiny 
of each one of these attitudes ? 

With regard to the first type, we believe that though 
partly rooted in the child's instinctive reactions, it is 
fashioned primarily by adult constraint. We shall have to 
analyse very closely this super-position of social influences 
on the spontaneous attitude of the individual if we wish to 
gain a more exact understanding of the idea of expiation. 

Among the instinctive tendencies must above all be 
mentioned the vindictive tendencies and compassion. 
For both develop independently of adult pressure. De- 
fensive and aggressive reactions are sufficient to explain 
how the individual, from at first inflicting pain upon 
his adversary in self-defence comes to make him suffer 
in response to all offences. Vengeance is thus con- 
temporary with the earliest defensive manifestations. It 
is very difficult to say, for example, whether the fit of 
rage of a baby of a few months old merely expresses the 
need to resist unwelcome treatment, or whether it already 
contains an element of revenge. At any rate, as soon as 
blows appear (and they do so at an extraordinarily early 
date, independently of any adult influence) it would be 

that the youngest pupils (4-6) can see nothing but expiatory punish- 
ment in the measures of reciprocity. The latter are not understood 
until about 7-8 on the average. 


hard to say where fighting ended and revenge began. 
Now, as Mme Antipoff in a short study on compassion l 
has very well shown, vindictive tendencies admit of being 
" polarized " very soon, under the influence of sympathy. 
Owing to its astonishing faculty for introjection and affec- 
tive identification the child suffers with him who suffers, he 
feels that he must avenge the unfortunate as well as himself, 
and experiences " vindictive joy " at seeing any sort of 
pain inflicted upon the author of other people's sufferings. 

But it is going a little too far simply to base the sense of 
justice on such reactions and to speak, as does Mme Anti- 
poff, of " an innate and instinctive moral manifestation 
which, in order to develop, really requires neither pre- 
liminary experience nor socialization amongst other chil- 
dren ". In order to prove her thesis, Mme Antipoff lays 
stress upon the fact that the vindictive tendencies become 
directly polarized upon the transgressor. " We have here," 
she concludes, " an inclusive affective perception, an 
elementary moral ' structure ' which the child seems to 
possess very early and which enables him to grasp simul- 
taneously evil and its cause, innocence and guilt. We 
may say that what we have here is an affective perception 
of justice" We may mention at once that nothing in the 
very interesting observations quoted by Mme Antipoff 
goes to show this innateness. She deals with observations 
on the behaviour of children between 3 and 9, and it is 
obvious that at the age of three, a child has already come 
under all sorts of adult influences such as can account for 
the fact that its polarization is now only in terms of good 
and evil. The proof of this is that the child speaks ; it 
says " serves him right " and " naughty boy ", etc. How 
could it have learned these words without coming under 
the moral influence of the person who taught them to it, 
and without accepting at the same time a whole set of 
explicit or implicit commandments? In a general way, the 
problem may be stated as follows. How can the vin- 

1 H. Antipoff, " Observations sur la compassion et le sens de la 
justice chez 1'enfant ", Arch, de Psychol., t. XXI, p. 208 (1928), 


dictive tendencies, even If they are polarized under the 
influence of compassion, give birth to the need for rewards 
and punishments and to retributive justice, unless the 
relations between individuals intervene to " regulate >J 
tins polarization, diminishing what is arbitrary and indi- 
vidual in the name of a normative element of eittier 
authority or reciprocity ? 

To our mind, when a child merely avenges some un- 
fortunate for whom he feels immediate compassion, neither 
the sense of justice nor the idea of punishment is yet at 
work. All we have is an extension of the vindictive 
tendency. But even if this sort of disinterested vengeance 
is a necessary, it is not a sufficient condition for the 
development of justice, disinterested vengeance will only 
become a " just " punishment when rules come in and 
make precise the distinction between what is right and 
what is wrong. So long as there are no rules, revenge, 
even disinterested revenge, will rest only upon individual 
sympathy of antipathy and will thus remain arbitrary : 
the child will not have the feeling of punishing the guilty 
and defending the innocent, but simply of fighting an 
enemy and defending a friend. But as soon, on the 
contrary, as there are rules (and they appear very early 
the boy of 3 observed by Mme. Antipoff is already satur- 
ated with them), as soon as we have rules, we get judg- 
ments of guilt and innocence and we get the moral 
" structure " of retributive justice. Where, then, do these 
rules come from ? 

Even if adults never interfered, the social relations 
subsisting between children would perhaps be sufficient 
to create them. The play of sympathy and antipathy is a 
sufficient cause for practical reason to become conscious 
of reciprocity. And the fact that the law of reciprocity 
leads to a certain type of punishment has, we believe, 
been sufficiently established in the preceding analyses. 
But in that case, the idea of expiation would never 
arise : mere vengeance would remain a private affair until 
such time as it would be considered immoral, and 


punishment by reciprocity would alone be held to be 

But the adult intervenes. He imposes commands which 
give rise to rules that are regarded as sacred. Disinter- 
ested vengeance, once it has been " polarized " by these 
rules, becomes expiatory punishment, and the first type 
of retributive justice is constituted in this way. When 
the adult is angry because the laws he has laid down are 
not observed, this anger is held to be just, because of the 
unilateral respect of which older people are the object and 
because of the sacred character of the law laid down. 
When adult anger finds vent in chastisement, this ven- 
geance from above appears as a legitimate punishment, 
and the resultant suffering as a " just " expiation. The 
idea of expiatory punishment is thus, taken as a whole, 
due to the conjunction of two influences : the individual 
influence, which is the desire for vengeance, including 
derivative and disinterested vengeance, and the social 
factor, which is adult authority imposing respect for given 
orders and respect for vengeance in cases where these 
orders are disobeyed. In a word, expiatory punishment 
is, from the child's point of view, revenge that may be 
likened to disinterested revenge (because it avenges the 
law itself) and that emanates from the authors of the law. 

How, then, are we to explain the passage from the first 
to the second type of retributive justice ? If the above 
remarks are correct, this evolution is nothing but a special 
case of the general evolution from unilateral to mutual 
respect. Since in every domain we have studied up till 
now, respect for the adult or at any rate a certain way 
of respecting the adult diminishes in favour of the 
relations of equality and reciprocity between children (and 
so far as is possible, between children and adults), it is 
perfectly normal that in the domain of retribution the 
effects of unilateral respect should tend to diminish with age. 
That is why the idea of expiation loses more and more of 
its power, and it is why punishments tend more and more 
to be ruled by the law of reciprocity alone. So that what 


remains of the idea of retribution is the notion, not that 
one must compensate for the offence by a proportionate 
suffering, but that one must make the offender realize, by 
means of measures appropriate to the fault itself, in what 
way he has broken the bond of solidarity. The situation 
can be expressed by saying that distributive justice (the 
idea of equality) definitely takes precedence over retribu- 
tive justice, whereas in the beginning the converse was the 
case. We shall come to the same conclusion in paragraph 
4 of this chapter. Let us add, finally, that the idea of 
reciprocity, often taken at first as a sort of legalized 
vengeance or law of retaliation expressed in quasi- 
mathematical form, tends of itself towards a morality of 
forgiveness and understanding. As we shall see later on, 
the time comes when the child realizes that there can be 
reciprocity only in well-doing. We have here a sort of 
repercussion of the form of the moral law on its content. 
The law of reciprocity implies certain positive obligations 
in virtue of its very form. And this is why the child, 
once he has admitted the principle of punishment by 
reciprocity in the sphere of justice, often comes to feel 
that any material punitive element is unnecessary, even 
if it is " motivated ", the essential thing being to make 
the offender realize that his action was wrong, in so far as 
it was contrary to the rules of cooperation. 

We have neglected, so as to deal with it separately, a question 
which it may be useful to discuss in connection with 
retributive justice. Do children consider it just, in general 
or in cases where the offender is unknown, to punish the 
whole group to which he belongs ? The question has a 
double interest, educational and psycho-sociological. 

Educational, because collective punishment has long 
been resorted to in the class-room, and, in spite of the 
many protests that have been raised against this practice, 
it is still more widely used than is generally thought. A 
certain importance therefore attaches to the way in which 


this custom reacts upon the child's mind. And the matter 
interests us psychologically as well. The history of penal 
codes teaches us that responsibility was long considered to 
be collective and communicable. The date is compara- 
tively recent when responsibility became individualized, 
and the more primitive conception survives to this day 
in many religious beliefs. M. Fauconnet, in the excellent 
book of which we have spoken, and which will occupy us 
again, has shown how the notion of communicable re- 
sponsibility is connected with that of objective responsi- 
bility. Now, as we claim to have shown above, the child 
believes in objective responsibility. Is there, then, a 
parallel and complementary tendency to conceive of 
responsibility as communicable ? 

In order to solve this problem, we gave the children a 
certain number of stories which lent themselves to conver- 
sation and which reproduced situations in which the 
question of collective responsibility usually arises. These 
situations seemed to us to be three in number : i The 
adult does not attempt to analyse individual guilt and 
punishes the whole group for the offence committed by 
one or two of its members. 2 The adult wants to dis- 
cover the transgressor, but the latter does not own up and 
the group refuses to denounce him. 3 The adult wants to 
discover the transgressor but the latter does not own up 
and the group is ignorant of his identity. In each of 
these three cases you ask the child whether it is fair to 
punish the whole group, and why. We examined about 
sixty subjects between 6 and 14, which was a sufficient 
number, considering the relative uniformity in the answers 
obtained. The children questioned are not the same as 
those with whom we were concerned in the last section. 

It should be noticed straight away that of the three 
situations contemplated, only the first is comparable to the 
situations that generate collective responsibility in primi- 
tive societies. But of course it is important to analyse the 
other two by way of corroborative proof. 

Here are the stories of which we made use : 


Story I. A mother tells her three boys that they mustn't 
play with the scissors while she is out. But, as soon as 
she is gone the first one says, " Let's play with the 
scissors ". Then the second boy goes to get some news- 
papers to cut out. The third one says, " No, Mother said 
we mustn't. I shan't touch the scissors." When the 
mother comes home, she sees all the bits of cut-up news- 
paper on the floor. So she sees that someone has been 
touching her scissors, and she punishes all three boys. 
Was that fair ? 

Story II. A lot of boys, as they were coining out of 
school, went to play in the street, and started throwing 
snowballs at each other. One of the boys throws his ball 
too far and breaks a window-pane. A gentleman comes 
out of the house and asks who did It. As no one answers 
he goes and complains to the school master. Next day 
the master asks the class who broke the window. But 
again, no one speaks. The boy who had done it says it 
wasn't he, and the others won't tell on him. What should 
the master do ? (If the child does not answer or misses 
the point, you can add details to make things clearer.) 
Ought he to punish no one, or the whole class ? 

Story III. Some boys are throwing snowballs against a 
wall. They were allowed to do this, but on condition they 
did not throw them too high, because right high up there 
was a window, and the window-panes might get broken. 
The boys had a lovely time, all except one who was rather 
clumsy and who was not very good at throwing snowballs. 
Then, when no one was looking he picked up a pebble 
and put snow all round it so as to make a good hard ball. 
Then he threw it, and it went so high that it struck the 
window, broke the window-pane, and fell into the room. 
When the father came home he saw what had happened. 
He even found the pebble with some melted snow on the 
floor. Then he was angry and asked who had done this. 
But the boy who had done it said it wasn't he, and so did 
the others. They did not know that he had put a pebble 
in his snowball What should the father have done, 
punished everyone, or no one ? 

Story IV. During a school outing, the teacher allowed 
the children to play in a barn, on condition that they put 
everything back as they found it before going away. One 
of them took a rake, another a spade, and they all went 
off in different directions. One of the boys took a wheel- 
barrow and went and played by himself, until he went 


and broke it. Then he came back when no one was 
looking and hid the barrow in the barn. In the evening 
when the teacher looked to see if everything was tidy he 
found the broken barrow and asked who had done it. 
But the boy who had done it said nothing, and the others 
didn't know who it was. What should have been done ? 
(Should the whole class be punished or no one ?) 

Guided by the requirements of the experiment, we 
invented other stories on the same themes, but there is no 
need to report them here as the results obtained were poor. 
It will be noticed that stories I and II correspond re- 
spectively to the first two situations we distinguished a 
moment ago, and stories III and IV to the third. 

With regard to the first of these situations we were 
unable in spite of our desire to do so l to discover in our 
children the least trace of collective responsibility. Little 
and big alike consider the mother of Story I to be unjust. 
You should punish each individual according to what he 
has done and not the group according to the misdeeds of 
one of its members. Here are some examples. 

RED (6) : " What do you think of that IThe one who 
hadn't touched them, he ought to have told. Was it fair or 
not fair to punish them all three ? No. Why ? Because 
there was one who hadn't done anything. How many 
should have been punished ? Two." 

STAN (6) repeats the story as follows : " Once there was 
a lady who was going shopping, and one of the boys took the 
scissors. The other one cut the paper, and the other one, he 
didn't do anything. In the evening she came back and she 
punished them all three. Is that fair ? She should have 
scolded two of them and not scolded the third." 

BOL (7) : " Once there was a mother who had three chil- 
dren. Then she went out to do some shopping, and then she 
told them not to touch the scissors. And then they did touch 

1 We have often been told that by questioning children with a little 
diplomacy you can make them say anything you like. Here is an 
example to the contrary. We hoped very much because we had 
counted upon it theoretically that the little ones at least would 
answer in conformity with the notion of collective responsibility. This 
hypothesis proved to be false and our desire was not sufficient to suggest 
it to the subjects who were questioned. 


them. Yes, who did ? The first and the second, but not 
the third. Yes, and then ? And then when the mother 
came home she saw that they had been touching the scissors, 
and then she punished them. Yes, how ? She sent them to 
led without any supper. That's right. Now, what do you 
think of this story ? It's pretty. Was it fair or not to 
punish all three ? No. Only the first two. Why not all 
three ? Because the third one hadn't disobeyed. And the 
other two ? Yes, they had disobeyed. And then ? They 
went to bed without any supper. Was it fair? Yes. 
They were three brothers. So two of them were punished. 
The third one didn't need to be punished ? No." 

SCRIB (9) : " The children oughtn't to have touched the 
scissors. She was quite right to punish them. Did she try 
to find out which one it was ? She punished all three of 
them. She ought to have asked who had taken the scissors. 
She says, ' Since no one will own up, I shall punish all the 
children '. // no one had owned up she ought to have 
punished them all three, but otherwise it wouldn't have been 
fair, and she ought to have punished only two of them since 
they'd have owned up." 

All the answers obtained are of this type. They show 
how utterly foreign to these judgments is the idea of the 
solidarity of the group as regards responsibility. This 
result is all the more remarkable when we consider that 
most children under 7, as we shall see later on, look upon 
everything that the adult does as " fair ". It is therefore 
in opposition to this tendency to justify the adult in all 
things that these children, in the particular case of our 
interrogatory, reject the idea of collective responsibility. 
It is true that the child is quicker to discover adult 
mistakes in the sphere of retributive than in that of 
distributive justice : a wrongly applied punishment seems 
to them more unjust than inequality of treatment. 

Scrib's spontaneous reflections on the expediency of 
punishing all the three children together if the first two 
had not owned up introduce us to situation II. Should 
the whole group be punished when the offender does not 
own up and when the innocent refuse to denounce him ? 

We are here in the presence of a problem which differs 
very widely from the classical question of collective re- 


sponsibility in the evolution of social groups, but yet 
throws a certain light upon it. For the mere fact of the 
offender refusing to own up and his comrades refusing to 
give him away establishes within the group a solidarity 
greatly superior to that which existed in it before. To the 
naturally given solidarity is added a solidarity that is 
willed and accepted by all. Are these conditions in which 
the child will admit the principle of collective responsi- 
bility ? 

If we base our statistics upon age we get only a very 
indeterminate result. At every age there are children who 
in Story II and similar tales consider that the whole group 
should be punished, and there are children who think that 
it would be more just to punish no one. Thus both types 
of answer were found to be characterized by the same 
average age (about 9, as we questioned subjects between 
6 and 12). But under this apparent homogeneity it is in 
fact possible to discern types of reaction that are quite 
distinct. According to children of one type generally the 
youngest everyone should be punished ; not because the 
solidarity of the group renders the responsibility collective, 
but because each is individually guilty, seeing that no one 
will show up the author of the offence and that it would 
be a duty towards the master to do so. According to 
children of a second type generally the older ones 
everyone should be punished, not because it is wrong not 
to " tell ", but because by the mere fact of having decided 
not to denounce the offender the class recognizes its 
solidarity. This is collective responsibility of a kind, but 
is willed by the individuals and not in itself compulsory. 
And finally there is a third type roughly speaking of an 
intermediate age according to whom no one should be 
punished; partly because it is right not to "tell", and 
partly because the guilty one is not known. It should be 
added that children of the first type, in addition to the 
argument reported above, think that everyone should 
be punished because a misdeed necessarily involves a 
punishment. By punishing everyone, justice is satisfied. 


Whereas children of the other two types regard the 
punishment of the innocent as more unjust than the 
impunity of the guilty. But these considerations appear 
far more clearly in connection with situation III, and we 
shall therefore not dwell upon them for the moment. 
Here are examples of the first type, 

RED (6), Story II : " And what did the master do ? 
He punished them all. Why all of them ? Because he 
didn't know who it was had broken the window-pane." 
" What did the one who broke the pane do ? He said 
they weren't to tell. And what did the others think about 
it ? That they oughtn't to tell. And did the others think 
that was fair ? Yes. What was ? Not to tell [This in 
Red's opinion is where the fault lies, one should 
have denounced the offender. Hence the collective pun- 
ishment.] But was it fair that they should all be pun- 
ished, or not fair ? It was fair. Why ? It wasn't known 
who did it. Have you ever been punished all together at 
home or at school ? No. We are asked [who is the offender] 
and then we say." 

BOL (7), story analogous to Story II : " What was he 
going to do ? Punish them. Who ? All four of them. 
Why ? The mother didn't know who had done it, so they all 
four had to be punished. Why ? Only one of them had 
thrown the snowball. Did the other three have to be 
punished too ? Yes. Why ? Because they wouldn't tell. 
And did the others think it was fair ? No. Why not ? 
Well, perhaps they did. Why ? Because they wouldn't 
tell. So they all four had to be punished.'' 

SCRIB (9), Story II : " What ought the master to have 
done ? He ought to find out. He did ask the others but 
they wouldn't tell him anything. They ought to have told. 
What would you have done ? I'd have told . . . because 
it's something wrong [to break a window-pane] that you 
mustn't do. . . . It better to tell,, because anyone 
who breaks a window ought to be punisiied. But these 
children didn't tell. What ought the master to do ? He 
must punish the whole class because no one has told. Which 
is fairest, to punish everyone or to punish no one ? 
Which is fairest ? To punish the whole class because no one 
would tell. They ought to be punished. Listen. That day 
one of the boys was absent because he was at home, ill. 
It was the day he wasn't there that the master said he 


would punish the whole class and keep everyone back for 
an hour on Thursday 2 . Now when Thursday came, the 
boy was well again. Should he have been punished like 
the others, or not ? He ought to be kept back ; everyone 
must go together. [It is true that] the whole class was not 
there when the window was broken [but] he must be punished 
too, as the whole class is being punished." This is about the 
most definite of the statements made by the children 
whom we questioned on the subject of collective responsi- 

HER (9), Story II : " What should be done ? Punish 
the whole lot of them. Would that be fair ? No, because 
the one who had done it said nothing, and he was the only one 
who ought to be punished. And should the others have 
told, or not ? Yes, they ought to have told, If you had 
been one of the others, would you have told or not ? I 
should have told the master. Would the others have 
thought that very nice of you ? No. And if the master 
had punished everyone, would that have been fair ? No. 
And what about punishing nobody ? Not that either. 
What should have been done ? Keep the whole class in for 
an hour. And would you have thought that fair ? I'd 
rather be punished even if they hadn't found the one who was 
guilty. Even if it wasn't you ? Yes" 

The two dominating ideas in these answers stand out 
clearly. On the one hand, there must be punishment, 
even if the innocent suffer ; on the other hand, no one is 
completely innocent since the whole class refuses to 
denounce the offender. Worth noting is Scrib's idea of 
the solidarity of the class which is such that even an 
absent pupil should be punished along with the others 
when he comes back. This is the dawn of collective 
responsibility properly so called. 

Here are examples of the second type : everyone should 
be punished because the class decides to stand together. 

SCHU (13) : " The whole class should be punished. 
Why p Because if no one owned up, someone would have 
to be punished. Why must someone be punished ? So 
that the one who broke the window should not have the whole 

1 Thursday Is the weekly whole holiday in French and Swiss schools. 



punishment. [Note the freely accepted solidarity.] Why ? 
Would you have thought it right to punish the whole 
class ? Because they ought not to have left only one to be 
punished ; it would have been mean to let him be punished. 
[Note this forceful formula.] And was it right of him not 
to own up ? No, it wasn't right of him. What would you 
have done ? I would have owned up. And the others ? 
They might have told on him. Why didn't they ? Because 
they get a thin time of it afterwards. [Their friends take 
them to task.] Why ? Because there are fellows who are 
pals. They don't say anything. Why ? So that he 
shouldn't be punished. Then what should be done ? The 
master should make the whole class pay for the broken window. 
And supposing the gentleman says, * I don't mind about 
having the window paid for ; all I want is that the boy 
who broke it should be punished ' ? Then they must try 
and find him or else punish the whole class. And if one of 
the boys was absent on that day, should he be punished 
along with the others when he comes back ? No, he 
mustn't. Why ? Because he was not one of the gang." 

SCHMO (n) thinks that " decent chaps " do not tell on 
a friend if he is likely to be punished. But it is up to the 
master to punish the whole class, since the guilty one does 
not reveal himself. " If you had been at that school 
would yon have thought it fair ? No, not quite, but if I 
had been the master it is what I would have done." 

This type clearly differs from the first. It is right not 
to tell (and it is not the refusal to denounce the offender 
which should be aimed at in punishing the whole class), 
but since the class by keeping silence joins forces with the 
offender, by doing so it declares war on the master who 
henceforth has the right to act with severity. From the 
master's point of view, then, the collective punishment is 
admissible though in itself it is neither obligatory nor even 

Here are examples of the third type of answer : the 
whole class should not be punished. 

HOT (7!) : " Which is the most fair, to punish everyone 
or no one ? To punish the one who did it. But they don't 
know who it was. Then it is most fair not to punish any- 
one, since they don't know who did it." " And supposing 


the master simply says that everyone must stay on after 
school until the one who did it owns up, would that be 
fair ? Yes, he is quite right then. That is different from 
before. And if everyone was kept on for two hours, would 
that be fair ? No." 

NIK (10) : " Some people have told me that the whole 
class ought to be punished, some that no one should be. 
What do you think IPunish no one. Why ? Because 
you don't know who it is. Is that quite fair or not ? I 
don't know. Is it the fairest thing that could be done or 
no t ? Yes, it is. Why ? Because it would mean punish- 
ing all the other children. To punish the whole class would 
be altogether unfair, would it ? No. Why not ? Be- 
cause then the one who did it would be punished too" 

For these children, clearly, individual responsibility 
alone comes into play: the essential thing is that the 
innocent should not be hit. It is more just, therefore, 
to punish no one. As for collective punishment, it is 
legitimate only in so far as it succeeds in reaching the 
offender himself. 

Of these few facts noted in connection with situation II 
we can therefore say that only those of the second type 
resemble collective responsibility. For the children of the 
first type do not think of the fault as in any way com- 
municable : if everyone is to be punished, it is because 
everyone is guilty, since the spectators of the misdeed 
refuse to denounce its author. The responsibility is 
therefore general and not collective. Only Scrib, in his 
desire to punish the absent pupil, marks a momentary 
exception to this rule, and anticipates the second type of 
answer. As to children of the third type, they are 
definitely hostile to ideas of communicable responsibility. 
So that only the second type remains, consisting curiously 
enough of the oldest children. But if in their view the 
group is responsible, it is because it wishes to be so and 
decides through solidarity to share the offender's punish- 
ment. Is this attitude comparable in any way to that of 
" primitive peoples " who consider the group to be con- 
taminated by the misdeed of one of its members ? Before 
deciding this point let us turn to situation III. 


The stories III and IV are those which enabled us to 
analyse the children's reactions in regard to these situa- 
tions. An individual misdeed has been committed, but 
the group does not know who is guilty : should it be 
punished as a whole or should no one be punished ? On 
this point the children's reactions were perfectly definite. 
According to the little ones, everyone should be punished, 
not because the group is responsible but because there 
must be a punishment at all costs, even if it strikes the 
innocent as well as the guilty. According to the older 
ones, on the contrary, no one should be punished because 
the penalty inflicted upon the innocent is more unjust 
than the impunity of the guilty. The older children are 
unanimous, at any rate from the age of 8-9, in saying that 
collective punishment is less just in the present situation 
than in the case of situation II. 

Here are examples of the reactions shown by the 
youngest children. 

MAR (6), Story IV : " What should have been done ? 
Punish the little boy. Did they know who it was ? No. 
Well then ? Take a boy and punish him. Any boy ? 
No, you would change [taking them in turn]." 

FRIG (6), Story IV : " What should be done ? Punish. 
Punish how ? In a dark room. Who ? The one who 
broke the cart. Did they know who it was ? No. Then 
what could they do ? Put them all in a dark room. What 
did the others say ? That it wasn't me who did it. What 
did they think then ? That they shouldn't be put [in the 
dark room]. If you were the teacher what would you 
do ? Put them all in a dark room. 1 ' 

VEL (6), Story IV : Everyone is to be punished. " Is 
it fair to punish everyone ? Yes, because he broke the little 
cart. Who do you mean by ' he ' ? The boy. Then is it 
fair to punish everyone ? Yes. They should all be kept in 
after school." 

Sxo (7), Story IV : " Then, should everyone be pun- 
ished, or no one ? / would punish half the class. And 
supposing you were in that half of the class, what would 
you say at being punished with the others ? / would 
think that it was fair." 

GRIB (9), Story III : " Everyone should give a little 


towards paying for the window-pane. Which would be 
fairest, that everyone should pay a little or that no one 
should pay anything ? Since it is someone in the class, 
everyone should give a penny." Story IV : " Since no one 
can say who it was, the whole class must be punished. The 
one who had broken it wouldn't say, so he [the master] said 
that everyone should be punished. 3 ' 

HER (9), Story IV : " If he [the master] couldn't find the 
one who broke it, he'd be better to punish the whole lot. But 
had the others seen him ? No Do you think it is fairer 
here in this story or in the first [Story II] ? It is fairer in 
the second story because no one saw who it was. Why did 
they not tell in the first story ? Because they didn't want 
to tell on him. There ought to be more punishing in the 
second one because no one knew who it was. Then why 
must you punish more if no one knows who it is ? 
Because [in Story IV] the others couldn't tell on him since 
titey didn't know. Were they right not to tell in the first 
story, or not ? They were right not to tell. [Her has 
therefore changed his mind since the first interrogatory.] 
Shouldn't they have told ?No." 

The children's reaction is unmistakable. They accept 
collective punishment, "but they do so, not because the 
group is responsible as a whole for the faults of one of its 
members, but simply because the guilty one is unknown, 
and that there must be a punishment at aU costs. The 
fundamental fact in these cases is not the feeling of soli- 
darity in the group, but of the necessity of punishment. 
Hence the curious answer given by Her, who deems it 
more just to punish everybody in this case than in situa- 
tion II. For in the last case the children act quite rightly 
in not denouncing the offender, and it would not be very 
fair to punish them, whereas if the offender remains 
unknown, there is nothing left to do but to chastise the 
whole group. 

We shall now give examples of the older children who 
regard collective punishment as unjust. 

DELLEN (9), Story IV : " What should be done ? Ask 
them who did & Punish them ? Yes. How ? Ask who 
took the cart. But he won't own up. Must one be taken 


at random and punished ? No. They don't know which 
one it- is who broke Ike cart. Should no one be punished ? 
Yes. . . . No, that still would not be fair because the one 
who broke it would not be punished. Punish everyone ? 
No. Only the boy who broke the cart. It would not be fair 
to punish everyone because the others didn't break it. 
Punish two or three of them ? No. No one at all ? 
Yes, that would be fairest. But you told me that the one 
who broke the cart would not be punished then. . . . 
Should everyone be punished ? No, because the others did 

NIK (10), Story III : " Some say that the whole class 
should be punished, some that no one should. TeE me 
what you think. No one should be punished. Why ? 
Because you don't know who it was. Do you think that is 
what would be fairest ? Yes. Why ? Because otherwise 
it would mean all the other children being punished.'' 
Story IV : "If you were the master what would you 
think was the fairest thing to do ? Not to punish anyone. 
When would you think it fairest to punish them all, in 
the story where they all saw the boy break the window- 
pane, or in this one ? In the first story. Why ? Because 
they know and they won't say. 13 

Even those among the older children who are attracted 
by the idea of collective punishment answer like Nik on 
the last point namely that collective punishment is less 
just where the offender is unknown to the group. For 
where everybody knows the author of the misdeed and 
refuses to denounce him, there is voluntary solidarity, as 
we saw in the case of situation IL But here there is 
complete independence between the individuals. The 
great majority of the older children therefore consider 
general punishment as a greater injustice than the im- 
punity of the guilty. 

It should be noted, however, that in some cases the 
child comes somewhat nearer to the notion of collective 
responsibility. This happens when the punishment chosen 
lends itself to this extension of the idea and seems to 
strike, not only at the offender, but at the child in general 
in its carelessness and its inferiority. For after all it 
seems hardly fair to keep everyone in for an hour when 


this might be done to the offender alone. But if the whole 
class is punished by being forbidden to borrow tools in 
future, this seems more reasonable. Such a measure 
strikes no longer at the innocent individual but at the 
child as such, the genus child like the genus humanum of 
the theologians. For since one member of the group has 
proved himself too clumsy to be allowed the use of a 
wheel-barrow, it is only normal that the group as a whole 
should be suspected of carelessness and that the responsi- 
bility should thus be extended to others. 

Nuss (7), Story IV : " What should be done to them ? 
They should have been told not to break the wheel-barrow. 
And did the master punish them or not ? Yes. How ? 
He told them not to touch things. Yes, but did he punish 
them ? Yes. He said they must never touch tools again. 
Did the children think it was fair ? Yes. Even those 
who hadn't done anything wrong ? Yes. 33 

These cases, along with that of Scrib (who wanted to 
have even the absent pupil punished), are those which 
come nearest to the classical idea of collective responsi- 
bility. And, as a matter of fact, even adults allow that a 
group may be penalized as when, for example, motorists or 
pedestrians are forbidden to go along a road where un- 
scrupulous individuals have been guilty of reckless be- 
haviour. But in this case, as in that of Nuss, the 
punishment is not expiatory ; it is a precautionary measure 
directed against individuals in general rather than against 
the whole group as such. 

On this point Mile A. M. Feldweg has been kind 
enough to supply us with a valuable complement to 
our enquiry. She questioned some forty children between 
5 and 13 by means of stories relating to situation I but 
which introduced into collective punishment the feature of 
being a general precautionary measure rather than an 
expiation properly so called. Here are two of these stories. 

Story V. There was a school with only two classes a 
class of big ones and a class of little ones. On Saturday 
afternoon, when no one was working very hard, the little 


ones asked the big ones to lend them one of their lovely 
animal books. The big ones lent it, telling them to take 
great care of it. But once, two of the little ones both 
wanted to turn over the pages at the same time. They 
quarreled and some of the pages of the book got torn. 
When the older ones saw that the book was torn they 
declared that they would never lend it to the little ones 
again. Were they right or not ? 

Story VI. A mother gave her three little boys a lovely 
box of coloured crayons and told them to be very careful 
not to drop them in case the leads got broken. But one 
of them who drew badly saw that his brothers drew 
better than he did, and out of spite (or " because this made 
him angry ") he threw all the crayons on the floor. When 
the mother saw this, she took the crayons away and never 
gave them back to the children again. Was she right to do 
this or not ? 

In contrast to Stories I and IV, these stories produced 
reactions that were apparently far more favourable to 
collective responsibility. For about half the children 
approved of the punishment in Story V and about one- 
fifth of that in Story VI. But if we try to find out the 
why of these judgments, we shall see immediately how 
totally foreign they are to truly collective responsibility. 

It is significant, to begin with, that the general punish- 
ment should receive a far greater measure of approval 
(about half) in the case of Story V than in that of Story VI 
(about one-fifth). For the punishment described in Story 
V consists far more of a preventive or protective measure 
than of a genuine punishment. The punishment in Story 
VI, on the contrary, contains an element of repression 
which renders it quasi-expiatory ; for the chalks in 
question were meant for the children from whom they 
were afterwards taken away, whereas in the first case we 
have to do with a book that had been lent and will now 
simply no longer be lent. 

On the other hand, the hesitation and even change of 
opinion shown by the subjects favourable to collective 
punishment are sufficient to show that their minds are not 
made up, and, above all, that the problem is new to them 


and does not correspond to an acquired or already ac- 
cepted idea. Finally, those answers that are definite 
hardly invoke anything more than the very idea of a 
preventive measure which we were emphasizing a moment 
ago. Here is an example of this type of answer. 

ROL (7 ; 6), Story V : " Were the older ones right or 
not ? They are right. Why ? They [the little ones] will 
tear it again, It was only one or two of them who tore it. 
Yes, but the others may tear it too. Could the book be 
given to the ones who had done nothing, or not ? Yes. 
Which would be fairest, to give it to them or not to ? 
Not to give it them, because they don't listen to what the big 
ones say to them. Suppose you were in the little ones' 
class, and had done nothing, would you think that fair, 
or not ? Yes, Why ? Because it was quite right. There's 
no need to go and tear a book that cost a lot! 1 

It is true that in certain rare cases an additional appeal 
seems to be made to the solidarity of the group as such. 
Here again, as in the case of Scrib (p. 237-8), one gets the 
impression that the child comes very near to collective 
responsibility. 'Here is one of these cases. 

HOCH (9), Story VI : " Was the mother right ? Yes. 
Why ? Because she's afraid they will throw them on the 
ground. Who do you mean by ' they ' ? Perhaps the 
others too. Suppose you have two brothers, and one of 
them throws the crayons on the floor ; is it fair or not that 
you and your other brother should not be able to draw 
any more ? Yes. But they have got no colours left becaitse 
of the brother. The other two ought to be allowed to draw. 
Will the third brother think that fair, or not ? No, 
because in the family they always lend each other the pencils. 
Then, what should be done ? Take the pencils away from 
all of them. 

Here, it will be seen, it is the unity of the group, felt as 
such, that impels towards collective punishment. But 
Hoch is obviously very undecided in his opinion. Besides, 
as we have said before, such views only appear in about 
one subject out of ten at the most. 

As to the other children, they are opposed to the idea 
of collective punishment even in regard to Stories V and 
VI. Here is the most interesting of the results obtained. 


HUF (13 ; 6), Story V : " It isn't right of them, Why ? 
They should take the book away [only] from those who tore 
it. If it was you who had torn it, would you think it 
fair that it should be given back to the others and not to 
you ? Yes, fair. Why not take it away from everyone ? 
The others might say it was not fair. 39 

Story VI : " Was the mother right ? No. What would 
you have done ? I'd have taken them away from the one 
who got angry. For always, or not ? For a certain time. 
And if it was you who had broken them, and no one was 
allowed to use the box, would you think that fair or not ? 
/ should think it was unfair. I should know that it was 
just to afflict [inflict] a punishment upon me. You wouldn't 
have thought it fairer that no one should be allowed to 
use the chalks ? Yes, while I was still in a temper, but not 
afterwards. I should know that I was the [only] one who 
deserved to have the box of chalks taken away from 

In short, Stories V and VI, like the others, are insuffi- 
cient to prove the existence of a spontaneous feeling for 
collective responsibility in the child. At the best, only in 
cases of very closely united groups, such as the family, 
does the child's judgment (as with Hoch) bear the fleeting 
impress of this collectivity. Broadly speaking, however, 
only such collective punishments are held to be just as 
can be regarded as preventive measures. 

Such, then, are the results of our enquiry. The sum 
total shows that in none of the three situations imagined 
is any judgment to be found comparable to the classical 
notion of collective responsibility. There are, at the most, 
little indications here and there, to which we shall return 
in a moment. In situations II and III, on the other hand, 
we can observe two consistent reactions, each of which, 
taken by itself, may be regarded as bearing upon com- 
municable responsibility. 

The first of these reactions is the belief in the absolute 
necessity for punishment. This belief can be observed in 
the younger children, and even leads them to demand the 
punishment of all rather than let the guilty one escape. 
Such an attitude is obviously necessary for the develop- 


merit of judgments of collective responsibility : before a 
whole group can be regarded as sharing the offender's 
guilt, the necessity of expiatory punishments must first be 
allowed. Fauconnet's contention seems to us irrefutable 
on this point. The emotion aroused by the crime is, he 
says, the source both of the collective reaction in which 
the punishment consists, and of the transference by 
contiguity and resemblance of the responsibility itself. 
In a sense, then, we may say with Fauconnet that re- 
sponsibility is born of punishment. But in the child this 
belief in the absolute necessity for expiatory punishment 
is not sufficient to liberate the judgment of collective 
responsibility. This is shown by our analysis of situation 
III it is the unknown offender and not the group as 
such that is aimed at by the collective punishment con- 
templated by the child. 

In the second place, we observed in situation II a sort 
of collective responsibility, but one that was voluntary 
and freely accepted : rather than denounce the offender, 
Ms comrades will declare their solidarity with him. Here 
again we come very near collective responsibility, but what 
this attitude lacks in order to be identified with the 
classical attitude is that the children should regard this 
solidarity as simply given and unavoidable. 

The problem may be stated as follows. Is the classical 
idea of collective responsibility, i.e. the necessity for the 
whole group to expiate the faults of one of its members 
more akin to the first or to the second of these reactions, 
to the absolute necessity for punishment or to the volun- 
tary solidarity of the group? The question is an im- 
portant one. As the first of these two attitudes is that of 
the younger children, whose morality is one of constraint 
(objective responsibility, expiatory punishment, etc.), and 
as the second is that of the older children, whose morality is 
one of cooperation (subjective responsibility, punishment 
by reciprocity, etc.), it is essential that we should de- 
termine whether a moral belief, which in the eyes of many 
passes as " primitive ", has taken its rise from one or 


other of these two ethics. Now, the results which we are 
trying to analyse here are doubly paradoxical. On the 
one hand, the only collective responsibility in which our 
children believe (the responsibility accepted by the group 
which wants to declare its solidarity) occurs among the 
older and not among the younger ones. On the other 
hand, the belief in obligatory expiation is strongest among 
the little ones and disappears precisely at the moment 
when this voluntary solidarity begins to develop. 

But everything becomes clear when we grasp the fact 
that collective responsibility in primitive societies pre- 
supposes the union of two conditions which in the child 
are always dissociated viz. a mystical belief in the ne- 
cessity for punishment and the feeling of unity and soli- 
darity within the group. The " primitive " is an adult 
living in organized societies. He may, under the influence 
of a gerontocracy, retain the essentials of a morality of 
constraint including the strictest ideas on the subject of 
retributive justice ; but in spite of this and simply owing 
to the fact of the powerful structure of the group to which 
he belongs, his feeling of the individual's participation in 
the collectivity is extremely strong. Responsibility is 
therefore collective at the same time as it is objective and 
as punishment is expiatory. In the child, on the contrary, 
we have to consider two phases. During the first, adult 
constraint develops the notions of objective responsibility, 
expiatory punishment, etc. The first condition for the 
existence of collective responsibility is therefore present. 
But the second condition is still lacking. For during this 
stage the child is essentially egocentric, and if he does 
have a feeling of close communion with the group (ego- 
centrism being by definition the confusion of the self with 
the not-self) it is primarily in connection with the adult 
or with the older child that this participation comes into 
play. There can therefore be no question of collective 
responsibility. During the second stage, on the contrary, 
the child enters more and more into the society of his equals, 
groups of which become organized in and out of school. 


Here, therefore, there is some possibility of collective 
responsibility, and the group does in point of fact declare 
its solidarity with the offender in cases where the latter 
enters into conflict with adult authority. But at the 
same^ stroke the first condition has ceased to be fulfilled ; 
the morality of cooperation has succeeded that of con- 
straint, and neither objective responsibility nor the belief in 
the necessity for expiatory punishment is any longer present. 
Hence we cannot speak of collective responsibility in the 
true sense. In our societies the child, as he grows up ? 
frees himself more and more from adult authority ; 
whereas in the lower grades of civilization puberty marks 
the beginning of an increasingly marked subjection of the 
individual to the elders and to the traditions of his tribe. 
And this is why collective responsibility seems to us to be 
missing from the moral make-up of the child, whereas it is 
a notion that is fundamental in the code of primitive ethics. 

3. " IMMANENT JUSTICE." A problem connected with 
that of punishment and one which we shall have to 
examine before passing on to the study of distributive 
justice is that of so-called immanent justice. If our 
hypotheses are correct, then the younger the child (not 
counting the first two years, of course) the stronger will be 
its belief in the soundness and universality of expiatory 
punishment, and this belief will give way before other 
values in so far as the morality of cooperation predomin- 
ates over that of constraint. During the early years of 
his life, the child must therefore affirm the existence of 
automatic punishments which emanate from things them- 
selves, while later, under the influence of circumstances 
which affect his moral growth, he probably abandons this 
belief. This is what we shall now endeavour to show. 

We told the children three stories in this connection. 

Story I. Once there were two children who were steal- 
ing apples in an orchard. Suddenly a policeman comes 
ilong and the two children run away. One of them is 
:aught. The other one, going home by a roundabout way, 


crosses a river on a rotten bridge and falls into the water. 
Now what do you think ? If he had not stolen the apples 
and had crossed the river on that rotten bridge all the 
same, would he also have fallen into the water ? 

Story II. In a class of very little children the teacher 
had forbidden them to. sharpen their pencils themselves. 
Once, when the teacher had her back turned, a little boy 
took the knife and was going to sharpen Ms pencil. But 
he cut his finger. If the teacher had allowed Mm to 
sharpen Ms pencil, would he have cut himself just the same? 

Story III. There was a little boy who disobeyed Ms 
mother. He took the scissors one day when he had been 
told not to. But he put them back in their place before 
Ms mother came home, and she never noticed anytMng. 
The next day he went for a walk and crossed a stream on 
a little bridge. But the plank was rotten. It gave way, 
and in he falls with a splash. Why did he fall into the 
water ? (And if he had not disobeyed would he have 
fallen in just the same ?) 

The first two questions were put by Mile Rambert to 
167 children from Geneva and the Vaudois Jura. (The 
same cMldren with whom we shall concern ourselves in 
the sequel on the subject of distributive justice ; these 
subjects have therefore not been questioned about punish- 
ments or communicable responsibility). We ourselves put 
question III and other analogous questions to children 
from NeucMtel. With regard to the first two, Mile Ram- 
bert was able to obtain statistics showing very clearly the 
influence of the child's mental age. Leaving aside the 
uncertain reactions wMch constitute about -Jth of the total, 
the answers affirming the existence of immanent justice, 
those, that is to say, where the subject maintained that if 
the child had not stolen or disobeyed he would not have 
fallen into the water or would not have cut himself, 
revealed the following percentages : 

AG 6 AGE 7-8 AGE 9-10 AGE 11-12 

86% 73% 54% 34% 

In addition to this it should be noted that in a class of 

backward children of 13-14 the proportion of answers of 

the same type was found to be 57%, which shows again 

that these answers are in inverse proportion to the mental 


age. Here are examples of this belief in the immanent 
justice of things. 

DEP (6), Story I : " What do you think of this story ? 
It serves him right. He shouldn't have stolen. It serves 
him right. If he had not stolen the apples, would he have 
fallen into the water ? No." 

CHR (6), Story III : " Why did he fall in ? God made 
him, because he had touched the scissors. And if he hadn't 
done what was wrong ? Then the board would have held 
out. Why ? Because he didn't -touch [would not have - 
touched] the scissors." 

SA (6), Story I : " What do you think of that ? The 
one who got caught was sent to prison, the other was drowned. 
Was that fair ? Yes. Why ? Because he had been 
disobedient. If he hadn't been disobedient would he have 
fallen into the water ? No, because he hadn't [ = would 
not have] been disobedient." 

JEAN (6), Story II : He cut himself " because it was 
forbidden to touch the knife. And if it had not been for- 
bidden, would he also have cut himself ? No, because the 
mistress would have allowed it." 

GRA (6) : Same answers for Story I. " What hap- 
pened ? The bridge cracked. Why ? Because he had 
eaten the apples. If he had not eaten the apples, would 
he have fallen into the water ? No. Why ? Because the 
bridge would not have cracked." 

PAIL (7), Story I : " What do you think of that ? It's 
fair. It serves him right. Why ? Because he should not 
have stolen. If he had not stolen, would he have fallen 
into the water ? No. Why ? Because he would not have 
done wrong. Why did he fall in ? To punish him." 

SCA (7) : " What do you think about it IOh yes, I 
know. If we do anything, God punishes us. Who told you 
that ? Some children. I don't know if it's true." Story 
II : " It serves him right. You ought to obey teacher. And 
if the teacher had allowed it, would he have cut himself 
sharpening a pencil ? No. He wouldn't have cut himself if 
the teacher had allowed it." 

BOE (8), Story III : " What do you think ? It serves 
him right. You shouldn't disobey. And if ... etc. ? 
No, he wouldn't have fallen in, because he wouldn't have done 
anything. 9 ' 

PRES (9), Story I : " What do you think IHe was 
punished as much as the other one, and even more. And if 


he had not stolen the apples, would he have fallen Into 
the water as he was crossing the river ? No, because he 
would not have needed to be punished." 

THIS (10), Story I : " He was punished. Neither of them 
ought to have stolen. If he hadn't fallen into the water he 
would have been caught. And If he hadn't been caught ? 
He'd have fallen into the water. Otherwise he'd have gone 
on stealing" 

Dis (n), Story I. : "He had his punishment too. 
Was it fair ? Yes. And if he hadn't stolen the apples 
would he have fallen into the water ? No, because [in 
that case] he didn't have to be punished.' 1 

Here are some examples of children who no longer believe 
in immanent justice, at least not in the stories we told 
them. But this does not prevent this belief from attaching 
itself to other objects, by becoming displaced and gradually 

GROS (9), Story III : " Why did he fall in ? Because 
the plank was worn out. Was it because he had disobeyed ? 
No. If he had not disobeyed would he also have fallen 
in ? Yes. He'd have fallen in just the same. The plank 
was worn out. 1 ' 

FLEU (12), Story I : " And if he hadn't stolen the 
apples, would he have fallen in too ? (He laughs.) The 
bridge isn't supposed to know whether he has stolen the apples. 9 ' 

BAR (13) : " It was perhaps a coincidence. But the 
punishment was what he deserved" 

FRAN (13), Story I : "And if he had not stolen the 
apples, would he have fallen into the water ? Yes. If 
the bridge was going to give way, it would have given way 
just the same, since it was in bad repair" 

But between two groups of clear cases, we find a series 
of intermediate examples, which are very interesting from 
the point of view of child logic, and whose original feature 
consists in saying that the event mentioned in our stories 
is certainly a punishment, but that it would have happened 
in any case, eveiLif there had been no previous offence. 

SCHMA (61), Story III : " Why did he faU into the 
water ? Because he told a lie. And if he hadn't done 
that, would he have fallen in ? Yes, because the bridge was 
O ld t Then why did he fall in ? Because he disobeyed his 


mother. And if he had not disobeyed would he have 
fallen in just the same ? Yes. The bridge was old, after all. 
Then why would the boy have fallen in who had not 
disobeyed? Not because of that. Why? . . ." 

MERM (9), Story II : " What do you think about it ? 
It served the boy right who fell into the water. Why ? It 
was his punishment. And if he had not stolen any apples, 
would he have fallen in ? Yes, because the bridge was not 
firm. But then it wouldn't have been fair. He would not 
have done anything wrong.' 3 

VAT (10) : " He was punished for his wicked deed. And 
if ... etc. ? May be he might have fallen in." 

CAMP (u), Story I : " And if he had not stolen would 
he have fallen in ? Perhaps he would have if the bridge 
had been rotten. But perhaps it was that God punished him." 

The little ones obviously do not feel the contradiction. 
The case of Schma is typical of children under 7 ; it is 
agreed that the child falls into the water because he has 
disobeyed, but he would have fallen in even if he had not 
disobeyed. The older ones, on the contrary, feel the 
difficulty well enough, but they try to reconcile the two 
themes of immanent justice and mechanical chance. 

Before taking up the question once again from the point 
of view of moral psychology, it will be well to ask whether 
and in what manner the child tries to picture the mechan- 
ism of this justice immanent in things which he seems to 
believe in. Does he establish an immediate bond between 
the offence and the physical punishment, or does he seek 
to find intermediate links in the form, for example, of 
miracles or of some sort of artificialist causality ? 

We have sometimes asked this very question. The 
subjects who answer " God did it " should immediately be 
put aside. This is sure to be a learnt formula. Many 
parents take advantage of the least coincidence that may 
occur between the minor accidents of which the child is a 
victim and his acts of disobedience, and declare with 
conviction, " You see, God has punished you," etc. But 
apart from this adult intervention we do not think that 
the question of the " how " really exists for the child. 
Whatever may be the manner in which belief in immanent 


justice first takes root, It seems quite natural to the cMld 
that a fault should automatically bring about its own 
punishment. For nature, in the child's eyes, is not a 
system of blind forces regulated by mechanical laws 
operating on the principle of chance. Nature is a har- 
monious whole, obeying laws that are as much moral as 
physical and that are above all penetrated down to the 
least detail with an anthopomorphic or even egocentric 
finalism. It therefore seems quite natural to little children 
that night should come in order to put us to sleep, and 
that the act of going to bed is sufficient to set in motion 
that great black cloud that produces darkness. It seems 
quite natural to them that their movements should com- 
mand those of the heavenly bodies (the moon follows us 
in order to take care of us). In short, there is life and 
purpose in everything. Why then should not things be 
the accomplices of grown-ups in making sure that a 
punishment is inflicted where the parents' vigilance may 
have been evaded ? What difficulty should there be in a 
bridge giving way under a little thief, when everything in 
nature conspires to safeguard that Order, both moral and 
physical, of which the grown-up is both the author and the 
raison d'etre ? 

With the older children,. from the age of about 8, this 
mentality tends to disappear little by little. Belief in the 
justice immanent in things also diminishes, the two 
processes being no doubt correlative. But where this 
belief does survive in our children, it is only among the 
little ones that it entails any sort of enquiry as to the 
" how " of its execution. What happens here is something 
analogous to the adult's use of finalism. A semi-educated 
man may very well dismiss as " contrary to science " a 
theological explanation of the universe, and yet find no 
difficulty in accepting the notion that the sun is there to 
give us light. Thus finalism, although at first united to a 
more or less systematic artificialism, comes to outlive it in* 
the end and even as does every habitual conception 
to give the illusion of intelligibility. The idea of a justice 


immanent in things could not appear unaided in the mind 
of a child of twelve, but it may survive in such a 
mind, as indeed in that of many an adult, without for 
that matter creating problems or giving rise to difficulties. 
We have not, therefore, observed any spontaneous 
reflections concerning the causal mechanism of immanent 
justice. Such preoccupations only appear in those who 
no longer believe that the physical universe functions 
like a policeman. Thus Fleu (age 12) could remark to us 
jokingly that " the bridge was not supposed to know 
whether the boy had stolen apples ". The little ones do 
not ask themselves whether or not the bridge " knows " 
what happens : they act as though the bridge did know, 
or as though the mana which guides all things knew in 
place of the bridge ; but they do not formulate this 
belief. All the same, we may ask ourselves what they will 
answer if they are pressed for a more exact statement. 
On this point, as in the case of our questions on child 
artificialism and animism, the child does not hesitate to 
invent myths, which have, of course, no value as beliefs, 
but which are the indices of an immediate and inex- 
pressible connection established by himself. 

SE (6J) : "It wouldn't have happened if he hadn't picked 
apples. Did the bridge know what the boy had done ? 
No. Then why did it break ? Perhaps the thunder made 
the bridge break. And did the thunder know ? Perhaps 
God saw, and then he made thunder by scolding. That broke 
the bridge and he fell into the water" 

Cus (6) : " Did the bridge know he had stolen ? No, 
but it had seen." 

EUR (6) : " The bridge must have known, since it gave way 
and he was punished" 'And in Story II : " Did the knife 
know ? Yes, It heard what the teacher said since it was on 
the desk. And it said, ' since the boy is going to sharpen it, 
he'll cut himself:" 

AR (6) : " Did the bridge know ? Yes. How did it 
know ? It had seen." 

GEO (7 ; 10) : " And if he had not stolen any apples, 
would he have fallen into the water ? No. It was his 
punishment, because he had stolen apples. Did the bridge 


? ff 09 fatf it broke because there was a wind, and the 
wind knew." 

These answers must not, of course, be regarded as 
corresponding to beliefs. Only the last may perhaps be 
said to contain an element of spontaneous belief. For we 
have often had occasion to see (C.W. and C.C.) the in- 
telligent role which the child seems to assign to the wind. 
Most of these answers therefore simply indicate that the 
child finds quite natural the connection between the fault 
that has been committed and the physical phenomenon 
which serves as punishment. When the child is forced to 
make this connection explicit, he invents a story arti- 
ficialistic in one case, animistic in the other. But this way 
of reacting proves nothing more than that nature, in the 
child's eyes, is the adult's accomplice, no matter what 
methods she may employ in the process. 

Yet the intermediate answers quoted above raise a 
problem. Some children maintain that the bridge has 
given way as a punishment and that it would have given 
way even if the boy had not stolen the apples. This can 
be very simply explained if we remember that a form of 
causality to which the mind has attached itself during a 
given period (such as the physical and moral precausality 
of the child of 2 to 7) never disappears all at once but 
coexists for a time with the later types of explanation. 
The adult is familiar with these contradictions and justi- 
fies them with a cloak of words. It is only to be expected 
that they should appear all the more frequently in chil- 

Let us now come to the main point and ask what, from 
the point of view of moral psychology, is meant by the 
facts we have put on record. For this purpose it will 
be necessary to state more precisely how the belief in the 
immanent justice in things is born and how it passes 
away. The intellectual dement of this belief is, moreover, 
such as to facilitate the question of its origin. 

Three solutions as to the problem of origin offer them- 
selves for choice. The belief in immanent justice may be 


inborn, or it may be the direct result of parental teaching, 
or, again, it may be an indirect product of adult con- 
straint a product, therefore, to which the child mind will 
itself have contributed under the double aspect, intellectual 
and moral, of its nature. 

The first solution is highly improbable. It has, however, 
been claimed that onanism gives rise to spontaneous 
feelings of remorse and to auto-punishment in thought or 
in act, whence it would be possible to infer in individuals the 
general presence of a predisposition to see in the events of 
life the mark of immanent justice. And as a matter of fact 
it cannot be denied that we can observe in masturbators 
a systematic fear of the retribution residing in things 
not only the fear of the result their habits may have upon 
their health, the fear of making themselves stupid, etc., but 
also a tendency to interpret all the chance misfortunes 
of life as punishments intended by fate. But would all 
these attitudes develop in a child who had not acquired 
the experience of punishment from the world outside him ? 
We are ready to believe that such ideas arise independently 
of any direct adult instruction, and in children whose 
habits are hidden from those around them. But it is those 
persons around them who are, after all, the indirect cause 
of this belief in punishments emanating automatically 
from things, and facts of this kind seem to us to speak far 
more strongly in favour of the third solution than of the 
first. Too many taboos relating to sex are imposed from 
the first years of life for the child's most secret reactions in 
this domain to be regarded as really inborn. At any rate, 
in order to prove the absolutely spontaneous character 
of auto-punishment and of the beliefs connected with 
it, one would have to bring up a child in very special 
circumstances, if not outside social contact of any kind. 

As to the second solution, it contains, as we have already 
seen, a large proportion of truth. A great many children 
think that a fall or a cut constitute a punishment because 
their parents have said to them, " It serves you right ", 
or, " That will be a punishment for you ", or, " God made 


it happen", etc. And yet, even if these propositions 
explain the majority of cases, we do not think that they 
cover the whole ground. In other words, we believe that 
situations frequently occur in which the child quite spon- 
taneously considers an accident of which he is the victim 
as a punishment, and we believe that this happens with- 
out anything analogous having been suggested by the 
parents in other situations. According to this hypothesis, 
the child, having acquired, thanks to adult constraint, the 
habit of punishment, attributes spontaneously to nature 
the power of applying the same punishments. The third 
solution therefore seems to us to contain a part of the 

Apart from the facts relating to onanism, we may 
mention here several examples which show how easily 
such attitudes are adopted by the child. 

I. A well-known German-Swiss psychiatrist has told us 
that one of the most vivid of his childish memories is that 
of having been prevented from taking apples out of a 
basket by the cover of the basket unexpectedly closing 
over it. The basket was open and the child had put his 
hand inside it, without, as a matter of fact, regarding 
himself as a thief, when the cover suddenly fell on to his 
arm. He immediately felt that he was doing wrong, and 
simultaneously that he was being punished. No one else 
was present at the scene. 

II. Another memory. A child would often look for 
animals to add to his natural history collection. On days 
when he had anything to reproach himself with, he had 
the feeling that his bag was a bad one and that it was so 
because of his misdeeds. 

III. We have cited elsewhere (C.W., p. 149) the case of 
the deaf-mute d'Estrella studied by W. James, who 
associated the inoon with the punishments to which he 
was subjected. 

IV. We have also described (C.W., p. 101) the singular 
reactions of those children who* looked upon nightmares as 
punishments for the faults they had committed during the 

In these four observations, which could be added to 
indefinitely, it seems to us that the child's attitude takes 


shape without any direct influence on the part of the 
adult . The subject is the only person who knows what 
happens to Mm, and he takes good care not to talk about 
it to those around him. Of course one cannot prove that 
children have never heard their parents invoking imma- 
nent justice ; it may be that all parents do so. But the 
ease with which the child interprets everything in terms 
of immanent justice seems to indicate that we have here a 
tendency corresponding to his own mentality; and this 
is all we wished to establish. 

Belief in immanent justice originates therefore in a 
transference to things of feelings acquired under the 
influence of adult constraint. But this does not fully clear 
up yet the moral significance of the phenomenon. In 
order to understand it, we have still to ask how such 
beliefs disappear, or at any rate diminish in importance 
with the mental age of the child. For the progressive 
decrease of answers pointing to the existence of these 
beliefs as the child grows older is a result worthy of our 
attention. What are the factors that account for this 
diminution ? 

One might point simply to the child's growing experi- 
ence and to his intellectual advance. Experience shows 
that wickedness may go unpunished and virtue remain 
unrewarded. The greater the child's intellectual develop- 
ment, the more clearly will he see this. Such an explana- 
tion, though true up to a point, would be too simple if it 
were brought forward to the exclusion of all others. For 
it is not such an easy matter to be guided by experience. 
The more we analyse the conduct which consists in 
consulting facts the more delicate and complex an opera- 
tion does it appear. Not only does experience presuppose 
the active participation of intelligence, but in order to 
eliminate the affective factors which run the risk of 
falsifying our interpretations we also require a veritable 
ethic of thought, an ethic which can only come into being 
in certain individual or social situations. M. Levy-Bruhl 
has shown very well how impervious to experience are 


primitive societies where the vital beliefs of the whole com- 
munity are involved. And when we see how " primitives " 
contrive to justify a magical or mystical attitude after 
repeated failures, we cannot help being reminded of those 
of our contemporaries who can never learn from facts, 
To take only the case of immanent justice, how many 
simple souk still think that even in this life people's 
actions are the object of equitable rewards and punishments, 
and would rather assume some hidden fault to explain a 
neighbour's misfortune than admit the fortuitous char- 
acter in the trials that befall mankind. Or how often do not 
the more charitably minded invoke the mystery of destiny 
to defend universal justice at all costs rather than in- 
terpret events independently of any presupposition what- 
soever. It is obvious, then, that even for adults to accept 
or reject the hypothesis of immanent justice is a matter 
not of pure experience, of scientific observation, but of 
moral evaluation and of a general attitude. 

It is therefore not mere experience, but moral experiences 
of a certain sort that will guide the child in one direction 
or another. What experiences do we mean ? In the first 
place, we may assume that it is the discovery of the 
imperfection of adult justice. When, as is almost bound 
to happen, a child is submitted to unjust treatment by 
his parents or his teachers, he will be less inclined to 
believe in a universal and automatic justice. We may 
recall in this connection the crisis which M. Bovet has 
described in the sphere of filial piety and which is so 
important from the point of view of the evolution of 
beliefs. But this discovery of the inadequate character of 
adult justice is only one episode in the general movement 
which takes the child away from the morality of constraint 
and towards that of cooperation. 

It would seem to be to this general process and to its 
consequences in regard to the idea of retribution that we 
must, in the last resort, look for an explanation of the 
progressive disappearance of " immanent justice ". 


In the last three sections of this chapter we have been 
led to the conclusion that the importance of expiatory 
punishment seems to decrease as the child grows older 
and to the extent that adult constraint is replaced by 
cooperation. The time has now come for us to approach 
the study of the positive effects of cuoperation in the 
domain of justice, and to do this we shall first have to 
analyse the conflicts that may take place between dis- 
tributive or equalitarian justice and retributive justice. 
For it is easy, as we shall try to show, to assume that 
equalitarian ideas acquire their force from cooperation and 
thus constitute a form of justice which, while it does not 
contradict the more evolved forms of retributive justice 
(punishment by reciprocity is due to the progress of just 
such ideas), is yet opposed to the primitive forms of 
punishment, and even ends by giving the preference to 
equality whenever the latter is in conflict with retribution. 

And such conflicts are very frequent in the life of the 
child. It often happens that parents or teachers favour 
the obedient child at the expense of the others. Such 
inequality of treatment, which is fair from the retributive 
point of view, is unjust distributively. How, then, will the 
child judge it at the different ages he traverses ? To this 
end, we told our subjects three stories, asking each time 
whether or not it was fair to favour the well-behaved 
child. The difficulty of the interrogatory lies in the fact 
that two questions necessarily interfere with each other 
in such cases that of the severity of the adult (a question 
of degree) and that of the conflict between retribution and 
equality (a question of principle). The second alone is of 
any interest, but it is difficult to eliminate the first. We 
limited ourselves to varying the stories in the following 
manner. The first mentions no special fault and es- 
tablishes the conflict between retributive and distributive 
justice in the abstract; the second introduces only 
negligible faults and minor punishments; the third, 
finally, brings in a punishment which may strike the child 


as very severe. In spite of variations (which are marked 
of course by deviations in the average ages of the corre- 
sponding types of answer), the children's -reactions evolve 
according to a relatively constant law. With the little 
ones punishment outweighs equality, whereas with the 
older ones the opposite is the case. 

Here is the first story. " A mother had two little girls, 
one obedient, the other disobedient. The mother liked the 
obedient one best and gave her the biggest piece of cake. 
What do you think of that ? " According to Mile Ram~ 
bert's statistics 70% of the children of 6 to 9 approve of 
the mother, and only 40% of those from 10 to 13. These 
figures are of value, of course, only as a general indication. 

Here are some examples of children who put retributive 
justice above equality. 

BAR (6) : " It was fair. The other one was disobedient. 
But was it fair to give more to one than to the other ? 
Yes. She [the disobedient one] must always do what she is 


WAL (7) : " They both ought to have been given the same 
[if they were good]. And the naughty ought to have been 
given nothing. She just ought to have been good. 11 

Gis (7!) : " The mother was quite right ? Yes, because 
you must always obey your mother. Was it fair to give one 
of them more than the other ? Yes, otherwise she would 
be more and more disobedient, and the mother doesn't love us 
very much. She likes better those who do what she tells 
them: 1 

BE (7:9): " It was right. Was it fair ? Yes, so as to 
show the other one how she would love her if she obeyed so 
that she should become obedient." 

VER (8) : " She was right to reward the one who does what 
she is told. Was it fair ? Yes. If they had both been 
obedient she would have given them both the big piece." 

GRA (9 ; 4) : " It was fair. Why ? Because she did 
what she was told. The other one ought to be punished. 
Was it fair that the mother should love one better than 
the other ? Yes, because the other was disobedient" 

HERB (9 ; 10) : "It was fair because the most obedient 
ought to have the best things. When we are obedient, people 
give us the best things." 


PIT (9) : " It was fair, because those who are obedient 
deserve more things than those who are disobedient. Was it 
fair that she should not love them both the same ? Yes. 9 ' 

BA (10 ; 5), a girl who is first in her class and whose 
character is inclined to be what children call " goody- 
good " (Fr. "petit saint "). " Was what the mother did 
fair? SA^ was guile fair! [Shrugs her shoulders.] 
\Vhy ? Because sfie rewarded the one who obeyed her." 

DEA (n) : " The mother was quite right. Why ? 
Because she was obedient. Tfo other hadn't the right to have 
as much as the one who obeyed her" 

Here, now, are examples of children who set equality 
above retributive justice. 

MON (6), G. : " Is it fair ?No. Why not IShe aught 
to have given t^i both the^samt.~\Vhy ? . . .Was what 
the mother did fair ? No." 

Rj (7 ; 6) : " They ougJit both to be given some cake. 
Why ? Because if they aren't, it isn't fair." 

SCA (7 ; 6) repeats the story correctly and realizes that 
the measure involved is a repressive one. But he affirms : 
" It isn't fair. They should always have the pieces that are 
the same. It's like' at home, when there is a piece of cake 
that's bigger than the others I take it, and then my brother 
snatches it away from me" 

PA (8) : " Ttiey should be given the same amount." Pa 
realizes that one 'of the girls has been given less than the 
Bother " because she ougfti to have been good ", but he insists 
there should be equality of treatment. 

MER (9 ; 6), G. : " The (me -who was disobedient ought 
to obey, but the mother ought still to give her the same a$ 
the other. Why ? Because you can't have people being 
jealous. 19 

PRES (10 ; o) : " The mother ought to have loved the other 
one too and been kind to her, then perhaps she would have 
become more obedient. Is it fair to give more to the 
obedient one ? No." 

THE (10 ; 7), G. : " She ought to Iiave given them both the 
same amount. Why ? Because tJiey were her daughters, 
she ought to Jiave loved them both the same.'* 

SON (10 ; 7) : " // she gives more to the nicest one, it will 
make the other one worse. But wasn't it fair to give more 
to the most obedient one ? No. Why ? She shouldn't 
have been giving her everything just because she was nice." 


JAX (n) : " The mother was wrong. Why ? She ought 
to have given them each a piece of cake of the same size. 
Perhaps it was not her fault if she was disobedient. Perhaps 
it was her parents' fault. No. It was her fault. Still, she 
ought to have had the same piece of cake! 9 

Dis (n), G. : " She ought to have given them both the 
same. Why ? Because she will only get naughtier and 
naughtier. She'll be revenged against her sister. Why will 
she be revenged ? Because she'll only get a small piece of 
cake. Was it fair, what the mother did ? No, it wasn't 
fair: 9 

ERI (12 ; 5) : " She ought to love them both the' same, 
without any difference between them. She may love the 
obedient one best, but without showing it so as not to make 
anyone jealous." 

HOL (12 ; 5), G. : " Even though the other one was not 
obedient, she shouldn't have made any difference between the 
two. She should have punished her in another way. Why 
should there be no difference ? Children should be loved 
the same. Sometimes the other one is jealous." 

MAG (12 ; n), G. : " It wasn't fair. Perhaps it wasn't 
altogether the fault of the disobedient one. She ought to have 
been taught, and not only loved less. Otherwise, she'll grow 
more and more naughty" 

DEJ (13 ; 2) : " It wasn't fair. On the contrary, the 
mother ought to have been just the same with the other one. 
Then perhaps she would have behaved better. Perhaps [in 
this way] she was jealous and sillier than ever. Whatever 
your children are like, you must love them all the same" 

PORT (13 ; 10), G. : " It wasn't fair. The other one saw 
that she wasn't loved. She didn't take any trouble to im- 

The two types of answer stand out in clear contrast to 
one another. For the little ones, the necessity for punish- 
ment is so strong, that the question of equality does not 
even arise. For the older children, distributive justice 
outweighs retribution, even after consideration of all the 
relevant data. It is true that both types of answer are to 
be found at every age, though in varying proportions ; 
but it is only natural, given the multitude of possible 
influences, that on so delicate a point as this the moral 
judgment should evolve with less regularity than it does 


when occupied with simpler and more objective orders of 
ethical fact. In homes where punishment is meted out in 
large quantities and where rules are rigid and weigh 
heavily on the children, these, if they are not secretly 
rebellious, will continue for a long time to believe in the 
superiority of punishment over equality of treatment. In 
a large family where moral education depends more upon 
the contagion of example than upon constant parental 
supervision, the idea of equality will be able to develop 
much earlier. Thus there can be no question of clear-cut 
stages in moral psychology. All the more significant, 
therefore, is the evolution which we noted as taking place 
between the ages of 6 and 13, and of which the age of 9 
would seem to be the turning-point. For the difference 
between 70% and 40% is a remarkable one, especially if 
it be borne in mind that after 12-13 preference is given to 
punishment in only 25% of the cases examined. 

Before turning to the examination of the other two 
stories and trying to draw a lesson from them, let us pause 
once more to see how different is the attitude of children 
who set a premium on retribution from that of children 
who demand complete equality. The first do not attempt 
to understand the psychological context ; deeds and 
punishments are for them simply so much material to be 
brought into some sort of balance, and this kind of moral 
mechanics, this materialism of retributive justice, so 
closely akin to the moral realism we studied before, makes 
them insensible to the human side of the problem. 
Whereas most of the answers we have quoted as examples 
of the predominance of equalitarian tendencies show signs 
of a singularly delicate moral sense the mother's prefer- 
ence for the obedient child will discourage the other, will 
make it jealous, lead it to revolt, and 'so on. All the very 
sound remarks made by Pres, by Son, and by Eri are 
sufficient to show that the child is no longer preaching a 
little sermon, as do the upholders of punishment, but is 
simply trying to understand the situation, though natur- 
ally under the influence of what he has experienced or 


observed in Ms own life. It is in this sense, once again, 
that we can mark the contrast between cooperation, the 
source of mutual understanding, and constraint, the source 
of moral verbalism. Jax even goes so far as to think that 
children are not always disobedient through their own 
fault, but sometimes through that of their parents. The 
psychologist can only admire this expression of a point of 
view which the average adult still seems so little capable 
of adopting. In short, then, we may take it that children 
who put retributive justice above distributive are those 
who adopt the point of view of adult constraint, while 
those who put equality of treatment above punishment 
are those who, in their relations with other children, or 
more rarely, in the relations of mutual respect between 
themselves and adults, have learnt better to understand 
psychological situations and to judge according to norms 
of a new moral type. 

It may be noted in passing how completely the results 
of this interrogatory confirm our findings in Section i 
concerning children's views on the utility of punishment. 
According to many of the older subjects questioned, it was 
the child to whom the results of his actions had been 
explained who was least exposed to a moral relapse, and 
not the child who had been severely punished. Similarly, 
in the present question, systematic punishment appears 
harmful to all those of the subjects whose psychological 
insight has become sharpened in the course of family and 
social life. 

We now come to the second story. Its object is to 
facilitate the analysis of the same problem, but in con- 
nection with trivial deeds devoid of any moral importance. 
" One afternoon, on a holiday, a mother had taken her 
v children for a walk along the Rhdne. At four o'clock she 
gave each of them a roll. They all began to eat their rolls 
except the youngest, who was careless and let his fall into 
the water. What will the mother do ? Will she give him 
another one ? What will the older ones say ? " The 
answers may be of three types. Not to give him another 


roll (punishment) ; to give him another, so that everyone 
should have one (equality), or to give him another because 
he is small (equity, i.e. equality allowing for the circum- 
stances of each, in this particular case the differences of 
age). Mile Rambert obtained the following figures from 
the 167 children she questioned : 













Age 10-12 . 
Age 13-14 

We ourselves presented some other children with a 
variant of the same story framed so as to eliminate the 
factor of equity introduced by the difference of age. " A 
mother is on the lake in a little boat with her children. 
At four o'clock she gives them each a roll. One of the 
boys starts playing the fool at the end of the boat. He 
leans right over the boat and lets his roll fall in. What 
should be done to him ? Should he have nothing to eat, 
or should they each have given him a little piece of 
theirs ? " The figures here are from 6-8 years old, 57% 
for the punishment and 43% for equality ; and 25% for 
punishment as against 75% for equality between the 
years of 9 and 12. 

Here are examples of answers favouring punishment. 

VA (6 J) : " He mustn't be given any more, because he lei it 
fall in. What does the older brother say IHe wasn't 
pleased, because the little one had let his roll fall into the 
water. He said it was naughty. Would it have been fair to 
give him any more ? No. He shouldn't have let his drop** 

MON (6J), G. : "He mustn't be given any more. Why 
not ? Because the mother isn't pleased. And what would 
his elder sister have said ? That he should be given another 
roll" (Cf. the opposition between solidarity and retri- 
bution !) 

PAIL (7) : " He shouldn't be given another. He didn't 
need to let it drop. And what would the older ones have 
said if the little boy had been given another roll ? That it 
wasn't fair : ' He's let it drop into the water and you go 


and give Mm another one.' Was It right to give him 
another one ? No. He hadn't been good" 

DED (8) : ** She mustn't be given another one because she 
let it drop. What did the mother do ? She was going to 
scold her. What did her elder sisters think ? That it was 
quite fair became she had not been careful.'* 

WY (9) : " Mustn't give her any more. Why ? As a 
punishment. 33 

The answers favouring punishment in connection with 
the story of the boat are naturally of the same type. 
Here, however, are two other examples, which will show 
to what extent, in the case of those who put retribution 
before equality, the criterion remains heteronomous and 
dependent upon the will of the parents. 

SCHMA (7) : " He didn't need to go and play the fool at the 
end of the boat. It would teach him for another time. Then 
what should have been done ? Not to go shares. And if 
the mother said they were to go shares ? Then they must 
do as she tells them. But was it fair, or^ not ? Fair, 
because you have to do what your mother says.'* 

JUN (9) : " They shouldn't go shares because it was his 
fault. His brothers decided to go shares. Was that fair ? 
I don't know. Was it nice of them ? Yes, nice. Fair 
and nice ? M ore nice than fair. The mother tells them 
to go shares. Is it fair ? Yes, then it is fair" 

And here are cases of children for whom equality out- 
weighs the necessity for punishment. As a matter of fact, 
there are already to be found in the preceding group 
subjects who appealed to equality. The careless child 
must be punished, otherwise he will get two rolls, which is 
contrary to equality. But the preoccupation here is only 
derivative and attaches itself only to the punishment, 
whereas the following children are actuated primarily by a 
search for equality which they set above any sort of 

SCA (7) : " The little boy should be given another one 
because he is hungry. And what do the others say IHe 
must be given some bread because the big ones have some, so 
he ought to have some [too] ." 

Zi (8) : " He ought to be given some more, because htUe 
children are not very clever ; they don't know what they are 


doing. Was It fair to the older one or not ? It isn't fair 
that tHe older one should have his roll, and not the little one. 
The older one ought to have gone shares.' 3 

PER (n) : " He ought to be given some more, because it 
wasn't his fault that he let it drop, and it isn't fair that he 
should have less than the others!' 

XA (12) : " He ought to have been given a little back by 
taking away from the roll what the child had already eaten. 
And what did the others say ? If they were nice, they said, 

* He is getting as much more as we have *. If they were horrid, 

* Serve him right ! * " (Fr. " Tant pis pour lui J"). 

MEL (13), G. : " They should have divided up what the 
other children had left and given some to the little chap. Was 
it fair to give him any more ? Yes, but the child ought to 
have been more careful. What does ' fair ' mean ? It 
means equality among everyone." 

Here are a few more answers obtained from the story 
about the boat. This story differs from the preceding one 
in that the fault is more clearly marked in the child who 
loses his roll, and in that the child is not represented as 
the youngest. There should therefore be all the more 
chance of punishment winning as against equality. Yet 
the reactions are the same as in the previous case, and the 
need for equality is even, after the age of 7-8, set up in 
opposition to the punishment demanded by the adult. 

WAL (7) : " They ought to have gone shares. But the 
mother said, ' No. It serves him right. He played the 
fool. You mustn't go shares with him/ Was that fair ? 
It wasn't fair because he would have less than the others. 
And when people play the fool, isn't it fair that they 
should have less than the others ? . . . If you were the 
father what would you say ? That he should be given 
some more." 

ZEA (8) : " They ought to have gone shares with him. 
Was it more fair or only nice of them to share ? More 
fair. But the mother said, 'No, -it's his own fault'. 
Well, I would have gone shares. Even if the mother said 
no ? Yes, you ought to share." 

ROB (9) : " He ought to be given some more. But he 
played the fool. They ought to share. What had he 
done ? He was playing. Sometimes you lose pennies that 
way. That's a lot worse ! But the mother had said they 
were not to lean over the edge. What ought she to have 


said, that lie should have some more, or not ? That they 
ought not to go shares. And what would the other boys have 
said ? That they ought to share, because it wasn't fair" 

SCHMO (no): " They ought each to have given him a 
little piece. Would that have been fairer or only nicer ? 
It was nicer, and fairer too. And if the mother says they 
mustn't share ? They should obey but it wasn't fair." 

And now here are cases of children, who, though not 
possessing the requisite vocabulary, make the legal dis- 
tinction between equity and justice. According to these 
children, from the point of view of pure justice, the boy 
who drops his roll should be given nothing more, because 
he has had his share along with the others, and if he loses 
what is his, that concerns himself alone. But in addition 
to considerations of pure justice, the circumstances of the 
individual must be taken into account. The child is small, 
clumsy, and so on ; thus a kind of superior equality 
requkes that he should be given some more. This more 
subtle attitude is naturally only to be met in the children 
of 9 to 12. Before this, the child already has these 
feelings, but does not succeed in distinguishing them from 
those of justice pure and simple. Here are examples. 

DEP (9), G. : " He must be given some more. What did 
the older ones say ? It's not fair. You've given two to the 
kid and only one to us. And what would the mother have 
answered ? She's the youngest. You've got to be sensible." 

BRA (9) : " He shouldn't have let it drop. He mustn't 
have another one. But it would be more fair all the same 
that he should have some more, that they should give him 
another. Would it be more fair, or only just more kind ? 
More kind, because he didn't need to go and let it drop 
into the water!' 

CAMP (n), G. : "The little boy ought to have taken care. 
But then he was a little boy, so they might give him a little 
piece more. What did the others say ? They were jealous 
and said that they ought to be given a little piece more too. 
But the little one deserved to be given a little piece more. The 
older ones ought to have understood. Do you think it was 
fair to give him some more ? . . . Of course ! It was a 
shame for the little one. When you are little you don't 
understand what you are doing." 


Now for our tMrd story. We can put the matter briefly. 
The answers obtained entirely confirm t&ose given up to now. 
But as the punishment in this case is particularly severe, 
the feeling of equality is sooner in gaining the ascendant 
over the need for retribution. " There was once a family 
with a lot of boys. They all had holes in their shoes, so 
one day their father told them to take their shoes to the 
shoemaker to be mended. Only, as one of the brothers 
had been disobedient several days before, the father said, 
' You won't go to the shoemaker. You can keep your 
holes since you have been disobedient/ " The children of 
6 to 7 were divided into 50% for equality, 50% for punish* 
ment. But after the age of 8, nearly nine-tenths are in 
favour of equality of treatment. 

Here are two examples of children who approve of such 
a punishment. 

NEU (7) : " What do you think about it ? It is fair. 
? Because he had disobeyed" 

FAL (7) : " It is fair. Why ? Because he was naughty. 
Is it fair or not that he should not have fresh soles to 
his shoes ? It is fair. If you were the father, would you 
take his shoes to be mended or not? / wouldn't take 

And a few examples of subjects who prize equality. 

ROB {9) : " It wasn't fair. The father had told them that 
they were all to be done" 

WALT (10) : " It wasn't fair. Why ? Because one boy 
would have a good pair of shoes and the other would have wet 
fed. But he had been disobedient. . . . What do you 
"think about it ? That it was not fair" 

Nus (10) : " It isn't fair . What should the father have 
done ? Punish him in some other way." 

Thus, whatever may be the variations in our stories, the 
answers are always the same. In case of conflict between 
retributive and distributive justice, the little ones always 
favour punishment, and the older ones equality of treat- 
ment. The result is the same whether we are dealing with 
definitely expiatory punishments, as in Stories I and III, 
or with a punishment that is the consequence of the act. 


as in Story II. We may note, moreover, without as yet try- 
ing to interpret it, the fact that here, as in the .case of moral 
realism, the child's reactions to the interrogatory i.e. Ms 
theoretical reflections are always a year or two behind 
his life reactions, that is to say, Ms effective moral feelings. 
A child of 7, say, who regards as just the punishments dealt 
with in our stories, would certainly feel their injustice if 
they were inflicted on him or on Ms friends. The interro- 
gatory therefore inevitably distorts the moral judgment. 
But here, as before, the question is to know whether the 
products of the interrogatory simply lag behind those of 
life or whether they do not correspond to anything that 
has ever been really experienced. As in the case of moral 
realism, we believe that it is chiefly a matter of lagging 
beMnd, and that our results do correspond to what can be 
observed in real life, though separated from this by a 
time-lag. Broadly speaking, then, we may say that if 
punishment holds the day during the early years, it 
gradually makes way for equality during the course of 
mental development. 

What is at the back of such an evolution ? It is obvious 
that equality will prevail over punishment by reciprocity, 
since the latter is derived from it. As to expiatory punish- 
ment we have notMng new to say on the subject. It is 
impossible to see how such a notion could have come into 
being except under the influence of adult constraint. There 
is nothing in the idea of right and wrong that implies reward 
or punishment. In other words, it is only because of ex- 
ternal associations that the altruistic or egoistic sentiments 
are bound up with expectation of rewards or punishments. 
And if this is so, whence can these associations arise if 
not from the fact that from its tenderest years the cMld's 
behaviour is submitted to the sanctions of adults ? 

But, tMs being so, how are we to account for the fact 
that retributive justice, wMch in all cases of conflict with 
distributive justice carries the day during the early years, 
should diminish in importance with the increase of years ? 
It can hardly be maintained that the fear of punishments 


is less strong at ten than it is at six. On the contrary, 
from the age of seven to eight onwards school punishments 
are added to those already incurred in the family, and 
even if punishments are less frequent at this age than at 
four or five, they have, on the other hand, a certain 
gravity which renders them the more apt to impress the 
youthful mind. So that the feeling of retributive justice 
ought really to increase with the years, and should be 
sufficiently strong to hold in check the desire for equality 
wherever this shows itself. Why is this not so ? 

Evidently because of the intervention of a new factor. 
The desire for equality, far from assuming an identical 
form at every age, seems, on the contrary, to grow more 
acute as the child's moral development proceeds. Two 
solutions are conceivable. It may be that equalitarianism, 
like retributive justice, is the fruit of the child's respect 
for the adult. Some parents are extremely scrupulous in 
matters of justice and instil in their children a keen sense 
of equality. Thus distributive justice may perhaps be 
merely a second aspect of adult constraint. But it may 
also be the case that, far from being the direct result of 
parental or scholastic pressure, the idea of equality de- 
velops essentially through children's reactions to each 
other and sometimes even at the adult's expense. It is 
very often the injustice one has had to endure that makes 
one take cognizance of the laws of equality. In any case 
it is hard to see how such a notion could take on any 
reality for a child before he had come in contact with his 
equals either in the home or at school. The relation 
between child and adult as such does not allow for 
equality. And since equalitarianism is born of the contact 
of children with one another, its development must at 
least keep pace with the progress of cooperation between 

We cannot as yet make our choice between these two 
hypotheses, for the analyses that will follow are essentially 
designed to facilitate this decision. But the facts we have 
presented already speak in favour of the second solution. 


For we have noticed that the champions of retributive 
justice are not generally those endowed with most psycho- 
logical penetration. They are apologists rather than 
moralists or psychologists. The votaries of equality, on 
the contrary, have given proof of a delicate sense of moral 
distinctions. And this feeling of theirs seems very often 
to be the product of reflections made on the moral clumsi- 
ness of the adult. In any case it is remarkable with what 
force these children contrast the cause of justice with the 
decrees of authority. But all this is only an impression, 
and we must now pursue our analysis of distributive 
justice and equality among children. 

5. EQUALITY AND AUTHORITY. The first point to be 
settled in an enquiry of this kind is the form in which the 
possible conflicts between the sense of justice and adult 
authority present themselves and the relation in which 
they stand to the subject's age. If we appeal to our 
memories of childhood, we very often find as examples of 
injustices (apart, naturally, from cases of unmotivated 
punishments) inequalities of treatment on the part of our 
parents. For it is very difficult, when one is dividing a 
piece of work among a few children, or expressing one's 
affection or one's interest to each, to maintain a strict 
impartiality and to avoid hurting the feelings of the 
more sensitive. It happens particularly often that chil- 
dren experience, either continuously or in bouts, those 
" feelings of inferiority " on which Adler has laid so much 
stress, and which make the best of them jealous, in spite 
of themselves, of their brothers and sisters. The least 
mistake in the treatment of such sensitive children will 
keep alive in them a vague impression of injustice, with 
or without foundation. What, then, is going to happen 
when children are told in the crude schematic form that is 
indispensable to an interrogatory addressed to all, stories 
which pit the desire for equality against the fact of 
authority ? Will the subjects who are examined put the 
adult in the right, out of respect for authority (justice 


being in this case confused with the Law, even if the latter 
is unjust), or will they defend equality out of respect for 
an inner ideal, even if the latter is in opposition with 
obedience ? As might have been expected from the pre- 
ceding results we found a predominance of the first solution 
among the little ones, and as the age of the subjects 
increased, a definite progression in the direction of the 
We made use of the four following stories : 

Story I. Once there was a camp of Boy Scouts (or Girl 
Guides). Each one had to do his bit to help with the work 
and leave things tidy. One had to do the shopping, 
another washed up, another brought in wood or swept the 
floor. One day there was no bread and the one who did 
the shopping had already gone. So the Scoutmaster 
asked one of the Scouts who had already done his job to 
go and fetch the bread. What did he do ? 

Story II. One Thursday afternoon, a mother asked her 
little girl and boy to help her about the house, because 
she was tired. The girl was to dry the plates and the boy 
was to fetch in some wood. But the little boy (or girl) 
went and played in the street. So the mother asked the 
other one to do all the work. What did he say ? 

Story III. Once there was a family with three brothers, 
one older one and two who were twins. 1 They all used to 
black their boots every morning. One day the big one 
was ill. So the mother asked one of the others to black 
his boots as well as his own. What do you think of that ? 

Story IV. A father had two boys. One of them always 
grumbled when he was sent messages. The other one 
didn't like being sent either, but he always went without 
saying a word. So the father used to send the boy who 
didn't grumble on messages oftener than the other one. 
What do you think of that ? 

Although we attach no magical value to figures, it may 
be of interest to mention here those obtained by Mile Ram- 
bert on some 150 children of 6 to 12 from Geneva and the 
Canton of Vaud by means of Stories I and II. The 
regularity of these results shows that we have at any rate 
to do with a form of evolution that proceeds as a function 

1 This detail is added so as to eliminate the question of age which 
several subjects spontaneously introduced. 


of age. The little ones incline to authority and even think 
the command given to the child quite just (not only 
should one obey, but the action commanded is just in 
itself, in so far as it conforms with the order given), 
whereas the older children incline to equality and think 
the order described in the story unjust. 


Age Obedience Equality Obedience Equality 

o/ o/ o/ o/ 

/o /o % % 

6 . . 95 5 89 ii 

7 * 55 45 41-2 58-8 

8 . . 33-3 66-6 22-2 77-8 

9 16-6 83-4 o 100 

10 . . 10 90 5 -9 94*i 

11 . . 5 95 o 100 

12 . * O IOO IOO 

For our part, we found at Neuch&tel, by means of 
Stories III and IV, that about 75% of the children of 5 
to 7 defend obedience, and that about 80% of the subjects 
between 8 and 12 defend equality. But let us leave these 
figures and turn to the qualitative analysis which is alone 
of a nature to tell us what the child is trying to say and 
whether he knows what he is thinking about. 

Four types of answers can be observed. First of all 
there are the children who regard the adult's order as 
"fair", and who thus do not distinguish what is just 
from what is simply in conformity with the order received 
or with the rule of obedience. Then there are the children 
who think the order unjust, but who deem that the rule 
of obedience comes before justice, so that it behoves us 
to carry out the order without comment. Children of this 
type can therefore distinguish justice from obedience, but 
think it evident that the latter must prevail over the 
former. In our statistics we have classed these two 
groups as one, seeing that they are linked together by all 
the intermediate cases. In the third place, there are the 
children who think the order unjust, and put justice 
above obedience. In the fourth place, finally, there are 


the children who also deem the order unfair, and do not 
necessarily regard blind obedience as incumbent upon the 
child in the story, but who think it better to be obliging 
and submit rather than argue and rebel. In the statistics 
we treated these two groups as one, owing to the autonomy 
given to the sense of justice in both cases. 

Here are examples of the first type, which naturally 
finds no representatives except among the very little ones. 

BAR (6J), G., Story I : " She ought to have gone to get the 
tread. Why ? Because she had been told to. Was it fair 
or not fair to have told her to go ? Yes, it was fair, because 
she had been told to." 

ZUR (6J), Story I : " He ought to have gone. Why ? 
To obey. Was it fair, what he had been asked to do ? 
Yes. It was his boss, his chief." Story II: "He should 
have gone. Why ? Because his sister was disobedient. He 
ought to be kind" 

HEP (7), G., Story I : " Was it fair what she was asked 
to do ? It was fair because she had to go. Even though 
it was not her job ? Yes, she had been told to go." Story 
II : " It was fair because her mother had told her to." 

ZIG (8), Story II : "He ought to have done both things 
because his brother wouldn't. Is it fair? It -is very fair. 
He is doing a good deed." Zig seems not to know the 
meaning of the word " fair ". But he has given us else- 
where an unequal division as an example of unfairness. 
In this Story II, therefore, what is just is identified with 
what is in conformity with obedience. 

JUN (9), Story III : " Was it fair IYes, I think so. 
What did the second boy say ? You ought to give three 
[boots] to one and three to the other. Very well then ? 
But you must do as the mother said. But was it fair, or 
was it because the mother said so ? It was fair ! " 

The character of these cases is obvious. It would be an 
exaggeration to say that the child of 6-7 has no idea of 
justice. Several of the above subjects hesitate to say 
outright that the orders given in the story were fair. 
Only, what is just is not differentiated in their minds 
from what is in conformity with authority, and it is only 
in so far as there is no conflict with authority that the 
idea of equality intervenes. So that with the little ones 
it goes without saying that an order received, even if it is 


contrary to equality, is still just, since it emanates from 
the adult. Justice is what is law. With the older of the 
children quoted above, this no longer goes without saying, 
but they decide that it must be so. 

Facts such as these lend confirmation to M. Bovet's I 
extremely interesting theory, according to which the child 
begins by attributing moral perfection to his parents and 
does not till the age of about 5-7 discover or face the fact 
of their possible imperfections. We shall return to this 
point. For the moment, the only question is whether 
such a systematic respect for the adult on the part of the 
child is of a nature to develop or to thwart the formation 
of equalitarian justice. With regard to the last set of 
answers examined, the hypothesis may be advanced that 
unilateral respect, which is of neutral content in relation 
to distributive justice parents can use the respect con- 
ferred upon them equally to uphold the example of justice 
and to impose a rale that is contrary to justice, such as 
the right of primogeniture that unilateral respect, then, 
does, by the very nature of its mechanism, constitute an 
obstacle to the free development of the sense of equality. 
Not only is there no possible equality between adults and 
children, but further, reciprocity between children cannot 
be produced to order. If it is imposed from without it 
leads only to a calculation of interests or remains subordin- 
ated to ideas of authority and external rales which are its 
very negation. According to the subjects whose answers 
we quoted above, what has been imposed is what seems 
just. It will be agreed that this is the very opposite of 
that autonomy required by the development of justice : 
justice has no meaning except as something that is above 

We shall now give examples from the second group of 
answers. The child always extols perfect obedience, but 
without complete inner acquiescence : authority still pre- 
vails over justice, but the two are no longer confused with 
one another. 

1 Bovet, op. df., The Child's Religion. 


CHRI (6), Story I: " Is it fair ? No, the girl does more. 
She'll be jealous. Did she go or not ? She went. Did she 
think it fair ? No, she'll say, ' It wasn't me was to go and 
fetch the bread '. Why did she go ? Because the chief 
wanted her to." 

DED (7), G., Story II : " She ought to have gone because 
her mother told her to. Was it fair ? No, because the other 
one ought to have gone." 

TRU (8 ; 7), Story II : " She ought only to have done one 
[one job]. Why ? It isn't fair if the boy didn't go [to 
fetch the wood]. But he didn't go, and so ? She ought to 
have done it att the same. Why ? To be obedient." 

HERB (9), G., Story II : " She ought to have gone at once. 
Why ? Because when you're asked to t you must go at 
once. Was it fair ? No, it wasn't her turn. Why did she 
go ? To do as she was told." 

Nuss (10), Story III ; " He ought to have done it, but it 
wasn't fair." 

WAL (10), Story III : " They should have blacked three 
shoes each. But the mother said one boy was to black 
two and the other four. Is that fair ? Not fair. The 
mother went out. What did the boys do as she had told 
them, or three boots each ? Three each. Was that right ? 
It would have been better to have done as the mother said. 
Was it fair? As [=what] the mother said, it wasn't 

REN (n), Story II : " He did it. Why ? You must 
obey. Was it fair ? No, not very." 

Thus these children, while they uphold the supremacy 
of obedience, distinguish between what is just and what is 
imposed by authority. Here are examples of the third 
group, of those, that is to say, who set justice above 

WAL (7J), Story II : " She shouldn't have gone because it 
wasn't her job. Why should she not have gone ? Because 
it was not her job. Was what the mother asked her to do 
fair ? Oh no. She shouldn't have gone. She should have 
done her work and the boy his. And if the mother asked 
her to ? . . . She would go. Why ? Because . . . she 
would have to." Thus Wai gives material constraint its 
due, but does not recognize any inner obligation. 

LAN (7 ; 6), Story I : " He shouldn't have done it because 
it wasn't his job. Was it fair or not to ask him to do it ? 


Not fair" Story II : " He ought not to have done it 
because the girl had gone away and it wasn't fair.'' 

PAI (8), Story I : "He said no, because it wasn't his job" 

DOL (8), Story III : " It wasn't fair. They should each 
have been given one boot. But the mother had said they 
must. The other one should have been given a boot. Ought 
they to do as they were told, or divide things equally ? 
They should have asked their mother." Story IV : " It's 
not fair. The father should have asked the other one too. 
But that is what the father had said, It's not fair. 
What should the boy have done ? Gone the messages, or 
not gone, or told the father to send the other boy ? He 
should have done nothing. Not gone." 

CLA (9 ; 8), G., Story II : " She ought to have done her 
own work and not the other ones. Why not ? It wasn't 
fair" Story I : " She oughtn't to have done it. It was not 
her job to do it. Was it fair to do it ? No, it was not fair." 

PER (10) : " He wouldn't go. He said the other one ought 
to go" 

FBI (ii), Story III : " He shouldn't do it. But the 
mother said he must. The mother was wrong. It's not 

SCHN (12), G., Story II : " She shouldn't have done it. If 3 
not fair that she should work twice as hard and not the other. 
What was to be done ? She should have said to her 
mother, ' It's not fair. I ought not to do double the work V 

For these children, unlike those of the second group, 
equality outweighs everything not only obedience, but 
even friendliness. The answers of the fourth type, on the 
contrary, present the special feature, that while he 
declares the order received to be unjust, the child thinks 
it should be carried out for the sake of being agreeable and 
helpful. Children of this group must not be confused with 
those of the second. For subjects of the second type 
consider that obedience comes before justice, whereas 
those whose reactions we are now going to examine 
recommend a voluntary mutual help which is superior 
both to bare justice and to forced obedience. The differ- 
ence is therefore considerable. On the one hand, justice is 
subordinated to obedience, thus to a heteronomous prin- 
ciple, on the other hand, justice itself is extended along a 


purely autonomous line of development into the higher 
form of reciprocity which we call " equity ", a relation 
based not on mere equality but on the real situation in 
which each individual may find himself. In this particular 
case, if strict justice is opposed to obedience, equity 
requires that the special relations of affection existing 
between parent and child should be taken into account. 
Thus a tedious job, even if it is unjust from the point of 
view of equality, becomes legitimate as a free manifestation 
of friendliness. Here are a few examples : 

PER (n ; 9), Story I : " He went to fetch some. Was it 
f a j r ? n wasn't fair, but it was obliging." 

BALT (n ; 9), Story II : " She did it. What did she 
think IThat her brother was not very nice.W&$ it fair 
that she should do this ? It wasn't fair, but she did U to 
help her mother." t( 

CHAP (12 ; 8) answers with regard to Story I that he 
thought his chief was a nuisance ", but in connection with 
Story II he says : " It depends on whether he's a good boy. 
If he is fond of his mother, he will do it ; otherwise he'll do 
the same as his sister so as not to have to work any harder 
than her." 

FED (12 ; 5) himself makes the distinction which seems 
to us to characterize the present type most clearly, and 
this in connection with Story I : " He must go and fetch 
the bread. What did he think IMy master orders me to; 
I must help him. Was it fair ? Yes, it was fair because it 
was from obedience. It wasn't quite fair if he was made to 
go, but if he accepted to, it was fair." One could not fonnu- 
late better the principle of autonomy which characterizes 
the attitude we are speaking of : if you are forced to do 
something against equality, it is unjust, but if _you accept 
to do a service, you are doing something superior to strict 
justice, and you are behaving with equity towards your 


GIL (12), Story II ; " He wasn't pleased. Did he do it ? 
_0h, yes. Was it fair ? No. Why did he do it ? To 
please his mother." 

FRI (12), G., Story II : " She might have refused. She 
thought that her brother would go and have a good time and 
that she would have to work. Was it fair or not to do it ? 
Not fair. Would you have done it or not ? I would have 
done it to please my mother." 


A law of evolution emerges sufficiently clearly from aH 
these answers. True, we cannot speak of stages properly 
so called, because it is extremely doubtful whether every 
child passes successively through the four attitudes we 
have just described. It is greatly a question of the kind 
of education the child has received. Thus the fourth 
kind of reaction might appear very early if one were 
willing to replace the absurd principle of blind obedience 
(" You've got to do it, and there's an end of it ! ") by an 
appeal to cooperation. A little girl of three of our ac- 
quaintance used to accept every suggestion from her 
mother, saying "I'll help you", where her pride would 
have resisted any sort of constraint. In addition to which, 
and in order to anticipate the inevitable objection, we repeat 
what we have said before that results of an interrogatory 
obviously come later in time than do those of real ex- 

With these reservations, however, it seems to us possible 
to distinguish three broad stages in the development of 
distributive justice in relation to adult authority. (And 
we shall see later on that the same holds good of the 
relations between children.) 

During the 'first stage, justice is not distinguished from 
the authority of law : " just " is what is commanded by the 
adult. It is naturally during this first stage that retribu- 
tive justice, as we saw in the last section, proves stronger 
than equality. This first stage might therefore be character- 
ized by the absence of the idea of distributive justice, 
since this notion implies a certain autonomy and a certain 
degree of liberation from adult authority. But there may 
well be something rather primitive in the relation of 
reciprocity, and the germs of equalitarianism may be 
present from the first in the relations that children have 
to each other. Only, so long as the respect for the adult 
predominates, that is to say, throughout the whole of 
this first stage, these germs could not give rise to any 
genuine manifestations except in so far as they created no 
conflict with authority. Thus a child of two or three 


years old will think it quite right that a cake should be 
equally divided between him and another child, or that 
he and a playmate should lend each other their toys. But 
if he is told that he must give more to the other child, or 
keep more for himself, he will promptly turn this into a 
duty or a right. It is unlikely, on the contrary, that such 
an attitude should survive for long in a normal child of 
ten or twelve ; the sense of justice here is founded on an 
autonomous feeling that is superior to any commands that 
may be received. 

During a second stage, equalitarianism grows in strength 
and comes to outweigh any other consideration. In cases 
of conflict, therefore, distributive justice is opposed to obedi- 
ence, to punishment, and very often even to those more 
subtle reasons that come to the fore in the third period. 

Finally, during a third stage, mere equalitarianism 
makes way for a more subtle conception of justice which 
we may call " equity ", and which consists in never 
defining equality without taking account of the way in 
which each individual is situated. In the domain of 
retributive justice, equity consists in determining what are 
the attenuating circumstances, and we have seen that this 
consideration enters very late into children's judgments. 
In the domain of distributive justice, equity consists in 
taking account of age, of previous services rendered, etc. ; 
in short, in establishing shades of equality. We shall come 
across fresh examples of this process in the course of the 
next section. 

Let us now turn to the analysis of more cases in which 
respect for authority enters into conflict with the sense of 
justice. It may happen not only that the child desires 
equality with his own kind, but that in some circum- 
stances he claims to be on a level with the adult himself. 
Mile Rambert conceived in this connection the happy idea 
of studying the situation to which the child is so often 
submitted, namely that of being made to wait at the 
counter of a shop while the grown-up customers are being 
served. She asked her subjects : " Is it fair to keep 


cMldren waiting in shops and to serve tie grown-ups 
first ? " The reaction was very definite. Only the very 
youngest of the subjects hesitated to say so, but the 
majority even of the six-year-olds maintained with 
astonishing precocity that each should be served in turn. 

Here are two examples showing respect for adult 

SAN (6|) : " Little children are not in such a hurry as 
grown-ups. 9 ' 

PAI (7!) : " Whoever comes first is served first. Have 
children as much right to be served as grown-ups ? No, 
because they are smaller and don't quite know how to give 
an order. Grown-up people have a lot to do and have to 
hurry" Pai adds that he is looking forward to growing 
up so as to " be able to give orders ". 

And a few examples of those who demand exact 

MART (9) : " They [the salesmen] ought not to keep 
children waiting. Why not ? Because it's not fair to keep 
them waiting. Grown-ups should always be served last [ = in 
turn]. Why ? Because sometimes little children are just as 
much in a hurry and it isn't fair [to keep them waiting], 
Should they be served when their turn comes or before 
the grown-up people ? When their turn comes." 

DEP (9) : " It isn't fair. Everyone should be served in 

BA (10) : " They ought to have served him [the child] 
when his turn came. Why ? Because it isn't fair to serve 
those who came afterwards." 

PRE (10) : " Even if he was little he shouldn't have been 
made to wait. He was shopping just as much as the grown-up 

It will be seen how definite is the desire for equality in 
these answers and how vividly they reflect the experiences 
of real life. 

To conclude our examination of the various contacts 
between authority and equality, let us try to analyse two 
school situations where the same factors may come into 
play : Why should one not cheat at school ? and : Should 


one " tell " if it is in the adult's interest or if the adult 
has commanded it ? 

Cheating is a defensive reaction which our educational 
systems seem to have wantonly called forth in the pupil. 
Instead of taking into account the child's deeper psycho- 
logical tendencies which urge him to work with others 
emulation being in no way opposed to cooperation our 
schools condemn the pupil to work in isolation and only 
make use of emulation to set one individual against 
another. This purely individualistic system of work, 
excellent no doubt if the aim of education be to give good 
marks and prepare the young for examinations, is nothing 
but a handicap to the formation of reasonable beings and 
good citizens. Taking the moral point of view only, one of 
two things is bound to happen. Either competition proves 
strongest, and each boy will try and curry favour with the 
master, regardless of his toiling neighbour who then, if he 
is defeated, resorts to cheating. Or else comradeship will 
win the day and the pupils will combine in organized 
cheating so as to offer a common resistance to scholastic 
constraint. The second of these two defence mechanisms 
appears chiefly in the older classes and, according to our 
personal memories, between the ages of 12 and 17. We 
hardly found a trace of it among the Elementary School 
children whom we examined. 1 In the first system the 
problem which arises is to know why cheating is reproved. 
Is it because the master forbids it, or because it is contrary 
to the equality between children ? 

Here again, the result of our enquiry is very definite. 
It shows a gradual diminution in the preoccupation with 

1 This may be because such confessions are neither easy to make 
nor to elicit. But as far as our personal recollections go, this cheating 
in common, though unconfessed, never seemed to us a sin. For years, 
as boys, we calmly did our home-work together and arranged to help 
each other in class within the limits of possibility. Nor was this 
clandestine work in common altogether useless, and we can recall 
many things that were learned by discussing them with our comrades. 
But of course this sort of thing diminishes individual effort to a great 
extent, a disadvantage from which precisely the work in common that 
is done in " Activity Schools " is free. 


authority and a correlative increase in the desire for 
equality. This result is all the more remarkable, because 
in this particular case authority and equality are not the 
only possible solutions. For the answers to the question, 
" Why must you not copy from your friend's book ? " can 
be classed under three heads : i " It is forbidden ", " it 
is naughty ", " it is deceit ", " a lie ", " you get pun- 
ished ", etc. We group all these answers under one head 
because, if the child's argumentation be analysed, the 
ultimate reason always proves to be the adult's prohibi- 
tion. It is naughty to cheat because it is deceiving, etc., 
and it is naughty to deceive because it is forbidden. 2 It 
is contrary to equality (it does harm to the friend, it is 
stealing from him, etc.). 3 It is useless (one learns 
nothing, one always gets caught, etc.). This third sort 
of answer is probably of adult origin : the child is merely 
repeating the sermon that has been preached to him 
when he has been caught cheating. It only appears after 
the age of 10. The percentages are : 5% at 10, 4% at u, 
and 25% at 12. Reasons in favour of authority are 
invoked in the following proportions : 100% at 6 and 7, 
80% at 8, 88% at 9, 68% at 10, 32% at n, and 15% at 12. 
The decrease is therefore unmistakable. The great ma- 
jority of the children simply say that cheating is forbidden. 
Only a small minority assimilates it to lying. Finally, 
equality is the reason defended by 16% of the children of 
8 and 9 years old, 26% of those of 10, and 62% of those 
of ii and 12. In the main, therefore, equality grows 
stronger with age, whereas the importance of adult pro- 
hibition decreases in proportion. 

Here are examples of answers that appeal to authority. 

MON (6i) : " Why must you not copy from your 
neighbour ? The master rows us." 
DEP (6J) : " Teacher punishes us/' 
TH (6J) : " Because it is naughty" 
MIR (6J) : " It's bad. You get punished: 1 

The definition " It is deceit " is given only by 5% of the 
children of 8 and 9 and by 10% of those of 10 to 12. 


MART (9) : " He shouldn't haw copied from his neighbour. 
He was being deceitful. Why must yon not copy ? 
Because it is deceit" 

Here are examples of children who appeal to equality. 

TH (9 ; 7) : " You ought to try and find out your self. It 
isn't fair they should both have the same marks. You ought 
to find out by yourself ." 

WILD (9 ; 4), G. : " If s stealing her work from her. 
And if the master doesn't know ? It's naughty because of 
the girl beside her. Why ? The girl beside her might have 
got it right [got a good mark] and her place is taken away. 9 ' 

Finally, let us quote a child to whom cheating is some- 
thing perfectly natural and in whose case the solidarity 
between children is clearly stronger than the desire for 

CAMP (n ; 10) : " What do you think about cheating ? 
For those who can't learn they ought to be allowed to have 
just a little look, but for those who can learn it isn't fair. A 
child copied his friend's sum. Was it fair ? He ought not 
to have copied. But if he was not clever it was more or less 
all right for him to do it" 

This last attitude seems to be rather the exception among 
the children we examined. But no doubt many others 
thought the same without having the courage to say so. 

If the letter only be considered in the answers appealing 
to equality, it might seem that competition was stronger 
in children than solidarity. But this is so only in appear- 
ance. In reality, equality grows with solidarity. This will 
appear from the study of one more question which we 
shall now analyse in order to obtain additional information 
on the conflicts between adult authority and equality or 
solidarity between children. We mean the question of 
" telling tales ". 

The contempt which every school child feels for tell- 
tales or sneaks (Fr. " mouchards ", " cafards ") the 
child's language is significant in itself and the spon- 
taneous judgment which is pronounced upon them are 
sufficient to show that this is a fundamental point in the 
ethics of childhood. Is it right to break the solidarity 


that holds between children in favour of adult authority ? 
Any adult with a spark of generosity in him will answer 
that it is not. But there are exceptions. There are 
masters and parents so utterly devoid of pedagogic sense 
as to encourage the child to tell tales. In such cases, 
should one obey the adult or respect the law of solidarity ? 
We put the question by laying the following story to the 
charge of a father whom we removed to a great distance 
both in time and space. 

" Once, long ago, and in a place very far away from 
here, there was a father who had two sons. One was very 
good and obedient. The other was a good sort, but he 
often did silly things. One day the father goes off on a 
journey and says to the first son : ' You must watch 
carefully to see what your brother does, and when I come 
back you shall tell me/ The father goes away and the 
brother goes and does something silly. When the father 
comes back he asks the first boy to tell him everything. 
What ought the boy to do ? " 

Here, again, the result was perfectly clear. The great 
majority of the little ones (nearly nine-tenths of those 
between 6 and 7) are of opinion that the father should be 
told everything. The majority of the older ones (over 8) 
ttgink that nothing should be told, and some even go so 
far as to prefer a lie to the betrayal of a brother. 

Here are examples of the different attitudes adopted, 
beginning with that of complete submission to authority. 

WAL (6) : " What should he have said IThat he [the 
other] had been naughty. Was it fair to say that, or not ? 
Fair. I know a little boy in the same story who said 
to his father : ' Look here, it's not my business what my 
brother has done, ask him himself.' Was he right to say 
this to his father ? Wasn't right. Why IHe ought to 
have told. Have you got a brother ? Yes. Then we'll 
pretend that you have made a blot in your copy-book at 
school. Your brother comes home and says : ' I say, 
Eric made a blot.' Was it right of him to say this ? He 
was right. Do you know what a tell-tale is ? It's telling 
what he [the other] has done. Is it telling tales if your 


brother says that you made a blot ? Yes. And in my 
story ? It's not telling tales. Why ? Because the father 
had asked him" 

SCHMO (6) : " He ought to say that he [the other one] was 
naughty. He ought to say what the other one had done. 
The father had told him to. The child answered his father : 
' Ask my brother yourself, daddy, I don't want to say/ 
Was it nice of him to say that, or not ? Not nice, because 
the father had asked him." 

DESA (6) : " He ought to have told. His father had asked 
him to. Should he have told, or not told ? He ought to 
have told. If he answers, * It's not my business what my 
brother does ', will that be all right ? He might have said 
that. Is it best to say that or to tell what the brother 
did ? Yes, it was better to tell what his brother did. Do 
you know what a tell-tale is ? No." 

SCHU (6) : " Should he have told, or not ? Yes, because 
Ms father told him to tell what his brother had done. Should 
he tell everything ? When they^ are very silly and naughty 
(Fr. f des grandes vilaines manieres ') you must tell every- 
thing. And when they are a little silly and naughty (Fr. 
' des petites vilaines manures ') ? No, because it's not very 
naughty. [This distinction anticipates the next stage !] 
Is that telling tales ? No, if you're asked to tell, it isn't 
idling tales. Might he have said, * Jean will tell you 
himself ' ? No. Or else, ' You ask Jean. It is not my 
business J ? No. Is it nice to tell tales about what your 
brother has done ? Yes" 

CONST (7), G. : " He ought to have told. The father had 
asked him to. Do you know what telling tales is ? It's 
telling things. Was it telling tales, or not ? It was telling 
tales. Have you any sisters ? Yes, one. She is eleven. 
Does she tell tales about what you do ? Yes. Tell me 
about once when she did it. Who did she tell tales to ? 
To mother. Tell me about it ? / didn't dare to go out. 
And I did go all the same. Was it nice to tell tales about 
that, or not ? Nice. Was she right to tell about it or 
not ? She was right" 

SCHMA (8) : " He ought to have told. Was it fair, or 
not ? Fair. Once he said that it wasn't his business. 
That was not fair, because his father had said he was to tell. 
Was he telling tales ? Just then he ought to tell because his 
father had asked him, but other times he ought not to tell 
because he hadn't been asked." 

IN (9) : " He ought to have told. I am going to tell you 


three stories : In the first, the boy did tell ; in the second, 
be told the father to ask Ms brother himself ; and in the 
third, he said that his brother hadn't done anything. 
Which was best ? The first. Why ? Because he told 
what he [the brother] had done, as his father had asked him 
to. Which way was the nicest ? The first. And the 
fairest ? The first one too. Do you know what telling 
tales is ? Telling what someone else has done. And here ? 
He didn't tell tales, he did what he was told" 

Here are cases of children who are opposed to telling 

TEHU (10 ; 6) : " / wouldn't have told the father because 
it was telling tales. I would have said, ' He's been good '. 
But if it isn't true ? I would have said, ' He's been good '. 
One child said, ' It's not my business. Ask him him- 
self. 1 Was that right ? I can't say that, it's not my 
business. I would have said he had been good" 

LA (7$) : " What do you think about it ? I wouldn't 
have told because the father would have spanked him. You 
would have said nothing ? No. I'd have said he hadn't 
done anything silly. And if the father asked you ? / 
should say that he hadn't done anything silly." 

FAL (8) : " Should he have told ? No, because that's 
telling tales. But the father had asked him to. He 
should have said nothing. Have said he'd been ni&* and 
good. Was it better to say nothing, not to answer, or to 
say that he'd been nice and good ? Say he'd been nice and 

BRA (9) : "It was rotten of him the one who went and told 
tales. But the father had asked him to. What should he 
have done ? Not told tales" 

MCHA (10) : " He should have said that he hadn't done 
anything. But the brother had played with his father's 
bicycle and burst one of the tyres. The father wouldn't 
be able to bicycle to his office the next day and would be 
late. All the same he shouldn't have told. [Then after 
some hesitation] He ought to tell so that he could put things 
right at once. 9 ' 

Here, finally, are two examples of subjects who hesitate. 
They are, as usual, the most illuminating, because they 
reveal the nature of the contradictory motives at the back 
of each of the two views of the matter. 


ROB (9) : "I don't know. Should the boy have told ? 
In a way it was fair, because the father had said so [asked 
for it]. Then what should be done IHe might have told 
the father a lie because [otherwise] it would have been telling 
tales. But he was bound to tell Which was most of a 
sport, the one who told what the brother had done, or the 
one who told a lie IThe one who didn't tell tales. And 
which would have been nicest ? The one who hadn't told 
tales. Which would have been most fair ? The one who 
told, because his father had said he must. 13 

WA (10 ; 3) : " He was quite right, because his father had 
told him to tett him [a pause, during which he hesitates]. 
Are you sure, or were you hesitating ? / was hesitating. 
Why? Because I was thinking that he might also say 
nothing, so that his brother shouldn't be punished. It's hard, 
isn't it ? Yes. Then, which one do you think is the most 
of a sport ? The one who said nothing. What would be 
the best thing to do ? It would be best for him to say 
nothing. What would he have said ? That he had been 

The mechanism of these judgments is clear. On the one 
hand there is law and authority : since you are asked to 
tell tales, it is fair to tell tales. On the other, there is the 
solidarity between children : it s wrong to betray an 
equal for the benefit of an adult, or at any rate it is 
illegitimate to interfere in your neighbour's business. The 
first attitude predominates among the younger children 
and is related to all the manifestations of respect for the 
adult which we studied before. The second prevails 
among the older children for reasons which have also been 
elucidated by all that has gone before. This second 
attitude is sometimes so strong that it leads the subject 
to justify lying as a means of defending a friend. 1 This 
interrogatory shows, even better than our previous results, 
the contrast of the two moralities that of authority and 
that of equalitarian solidarity. The style of speech used 

1 It should be noted in passing that this is a clear case of a lie being 
evaluated as a function of the motive that inspires it. The children 
who think it " sporting " (Fr. chic type) to tell a lie to protect a 
brother told us very definitely that the same lie would be " naughty " 
(Fr. vilain) if told in self-protection. 


by children in this connection is highly significant, and 
one may say that the terms used by children to describe 
behaviour in school are sufficient to differentiate the two 
types of reaction. The expression that symbolizes the first 
type most clearly is that of the "petit saint".' 1 The 
" petit saint " (lit. little saint) is the boy who ignores his 
playmates and only cultivates the master, who always 
sides with the grown-ups against the children. It is the 
well-behaved submissive pupil. This is how he is de- 
scribed by some children of 10 to 12 years old : " It's a 
chap who is always hanging on to his mother's skirts." 
" It's a ttche-cul (a groveller)/' " it's a Itche-cuteur (idem)" 
" a boy who tells tales," etc. The opposite of the " petit 
saint " is the " chic type " or sport, one who on occasions 
is up against the established order, but who is the incar- 
nation of solidarity and equity between children : " It's a 
fellow who will give aU he has to the others," " One who 
does not tell tales/' " A boy who plays again with the 
others when he has won everything at marbles/' " One 
who is fair," etc. 

This is a psychological work, and it is not for us to take 
up a moral standpoint. And yet when it comes to fore- 
telling character, it is perhaps worth while raising the 
question as to which of these two the " petit saint " or 
the " chic type " will develop into what is generally felt 
to be the best type of man and of the citizen ? Given our 
existing system of education, one may safely say that there 
is every chance of the " chic type " remaining one all his 
life and of the " petit saint " becoming a narrow-minded 
moralist whose principles will always predominate over his 
common humanity. 

The conclusion to be drawn from the above facts would 
therefore seem to be the following. Equalitarian justice 
develops with age at the expense of submission to adult 
authority, and in correlation with solidarity between 
children. Equalitarianism would therefore seem to come 

i English equivalents for these expressions will occur to the reader 
" goody-good ", " pi ". etc. [Trans.] 


from the habits of reciprocity peculiar to mutual respect 
rather than from the mechanism of duties that is founded 
upon unilateral respect. 

6. JUSTICE BETWEEN CHILDREN. If the result of our 
previous analyses is correct, the most favourable setting 
for the development of the idea of distributive justice and 
of the more advanced forms of retributive justice would 
seem to be the social relations between contemporaries. 
Expiatory punishment and the earlier forms of retributive 
punishment would seem to be created by the relations 
between children and adults. The time has now come for 
us to undertake a direct verification of these hypotheses 
by trying to see what is the child's conception of justice 
between comrades. Two points have to be considered 
punishment between children and equalitarianism. 

It cannot be denied that there are elements of retribu- 
tive justice in social life among children. The cheat is 
sent to Coventry, the fighter gets back the blov/s that he 
deals, etc. But the problem is to know whether these 
punishments are of the same kind as those to which the 
child is generally submitted by the adult. It seems to us 
that they are not. Adult punishment gives rise in the 
child's mind to ideas of expiation. A lie, or an act of 
insubordination, for example, will mean being shut up, or 
being deprived of a pleasure. To the child this punish- 
ment is a sort of putting things right, which will re- 
move the fault by placating authority. At any rate, the 
punishment is regarded as " fair " only in so far as the 
feeling for authority is present, together with remorse at 
having offended this authority. This is why, as the years 
go on and unilateral respect diminishes in strength, the 
number of punishments approved of by the child also 
grows sensibly less. As we saw at the beginning of this 
chapter, punishment by reciprocity gradually supplants 
expiatory punishment and in many cases even ends by 
being considered useless and harmful. Punishments be- 
tween children, on the contrary (except in the relations of 


older to younger children as in games with rules), could 
hardly rest on authority and consequently could not 
appeal to the idea of expiation. And we shall see, as a 
matter of fact, that they nearly all fall under what we 
have called punishment " by reciprocity " and are con- 
sidered " fair " in the measure that solidarity and the 
desire for equality among children is on the increase. 

It is possible to distinguish two (more or less arbitrary) 
classes of punishments among children. The first belongs 
essentially to games, and is applied when a player in- 
fringes upon one of the rales or customs. The other 
appears as chance determines, wherever the bad behaviour 
of some calls forth the vengeance of others, and where 
this vengeance is submitted to certain rules which render 
it legitimate. Now we shall find that not one of this last 
variety of punishment can be classified among expiatory 
punishments. When a boy gives blow for blow, etc., he is 
not seeking to chastise, but simply to show an exact 
reciprocity. We shall see, moreover, that the ideal aimed at 
is not to give more than one has received, but to mete out 
its mathematical equivalent. As for collective punishments, 
they are nearly all of the type " by reciprocity ", with one 
or two exceptions which we must now examine more closely. 

In the sphere of play, for example, we found only non- 
expiatory punishments. The boy who cheats is excluded 
from the game for a period of time proportionate to the 
gravity of his offence. Marbles unlawfully won are 
restored to their owner or distributed among the honest 
players. Similarly, where things are exchanged, if the 
strong takes advantage of the weak, he is brought to 
order by others stronger than himself, made to restore 
goods acquired by illicit bargains, and so on. In none 
of all this is there expiatory punishment properly so 
called, but only restitutive penalties, acts of exclusion 
marking a breach in the bond of solidarity, and so on. 
Only in cases of exceptional gravity, on the occasion of 
those crimes which Durkheim characterizes as directed 
against the " strong and definite feelings of the collective 


consciousness '*, have we noted the appearance of expia- 
tory punishment. For example, there is a certain board- 
ing school in NeucMtel where the " cafards " (sneaks) are 
ritually conducted " an jus " (to stew) ; which means 
that after school the whole body waits for them and 
leads them forcibly down to the edge of the lake, where 
they are ducked in the cold water with all their clothes 
on. But how is it that everyone regards this punishment 
as legitimate? There is obviously in each child the 
feeling of a moral authority presiding over executions of 
this kind. But is the authority that of the particular 
group taken at the moment when the event is happening ? 
Do the children who at a given moment constitute a class 
and are united by relations of reciprocity, do these chil- 
dren, by the mere fact of being so grouped, succeed in 
creating a collective consciousness, which will impose upon 
each and sundry its sacred character and thus become the 
equivalent of adult authority ? If this were the case the 
distinction between cooperation and constraint would 
become illusory : the union of a certain number of indi- 
viduals living in reciprocity with each other would be 
sufficient to produce the most rigid of constraints. But 
things are not so simple as this, and in the facts we are 
discussing there is a factor of age and tradition which 
makes the example in question comparable to those in 
which pressure is brought to bear upon the child by the 
adult. For putting an offender " to stew " is an ancient 
and venerable custom, and the class which for the time 
being is invested with the divine right of chastising the 
criminal is fully conscious of carrying on a time-honoured 
tradition. And it is our belief that because of this con- 
straint exercised by tradition the punishment seems just 
and becomes an expiation. I remember very definitely 
having been filled with two contradictory feelings when, 
as a school-boy, I was a witness for the first time of one 
of these sacred immersions. On the one hand, there was 
the feeling of the barbarity of the punishment (it was 
mid-winter), but on the other hand, there was the feeling 


of admiration and almost of respect for the " elders " of 
the class who could thus incarnate a part which we all 
knew to have been played by leaders of the older classes 
in similar circumstances. In short, the ducking of the 
sneaks, which was at first simply an act of vengeance and 
felt perhaps to be cruel by those who were not directly 
interested, had become for me, by being ritualized and 
transmitted from generation to generation, the expression 
of a just expiation. This shows that in the rare cases 
where punishments between children are truly expiatory, 
a factor of authority, of unilateral respect, of the con- 
straint of one generation upon the other has been intro- 
duced. Wherever this factor plays no part, punishments 
between children are and remain merely punishments " by 
reciprocity ". 

We now come to " private " punishments. Private 
punishment is, at its origin, revenge rendering evil for evil 
as one renders good for good. But can this vengeance be 
submitted to rules and to this extent appear legitimate ? 

We shall see that this is the case, and that this pro- 
gressive legitimacy is in direct ratio to the growth of 
equality and reciprocity among children. 

Mile Rambert put the two following questions to the 
167 children whom she interviewed : (I) " There was a 
big boy in a school once who was beating a smaller boy. 
The little one couldn't hit back because he wasn't strong 
enough. So one day during the recreation he hid the big 
boy's apple and roll in an old cupboard. What do you 
think of that ? " (II) " If anyone punches you, what do 
you do ? " 

The statistics show very clearly that reciprocity grows 
with age, and that in the same measure the punishment 
seems just. With regard to the first story, two answers 
are possible. " It was naughty ", or " The little one was 
quite right to pay him back ". The second answer was 
given in the following proportions : 

Age 6 Age 7 Age 8 Age 9 Age 10 Age n Age 12 

19% 33% 65% 72% 87% 91% 95% 


Here are examples of children who do not approve of 
the little boy. Curiously enough, they are thus mostly 
the little ones themselves. 

SAV (6) : " He oughtn't to have done it because it is 
naughty (Fr. mechant *). Why ? Because you get hungry, 
and then you look and you can't find it again. Why did 
the little chap take his roll ? Because the big one was 
naughty (mechant). Should he have taken it or not ? 
No, because it's naughty (m&hant)." 

PRA (6) : " He shouldn't have done it because it was the 
big boy's roll. Why did he do it ? Because the big boy 
was always beating the little one. Should he have let him 
beat him ? No. He ought to have defended himself. Not 
given in. Gone away. Why shouldn't he have taken the 
roll ? It's not fair to take things. You mustn't take 

MOR (6) : " He shouldn't have taken it. Why tIt's 
naughty (mechant). Why did he do it ? The other one 
beat him. Was it fair to take it ? Not fair. He ought to 
have told the teacher." 

BLI (6) : " He shouldn't have, because he was a thief. 
What should he have done ? Tell his mother. Should he 
have hit back ? No. His mother will scold him [the big 

DED (7) : " He shouldn't have done it because it wasn't 
nice of him. Why did he do it ? Because his brother was 
always hitting him. What should he have done ? Let the 
boy beat him and tell his mother. Not hit back." 

Ric (7 ; 6) : " He shouldn't have, because it's dis- 

TEA (8) : " He shouldn't have done it. Why ? After- 
wards the other boy looked for it everywhere and had nothing 
to eat. Why did he hide the roll ? Because the big one had 
beaten him. Was it fair, then ? No. Why ? He should 
have told the master." 

MAR (9 ; 8) : "He shouldn't have done it. Why ? 
Because it was stealing. But the other chap had beaten 
him. He should have told the master. Is it fair to take a 
revenge ? Yes . . . [hesitates] no." 

PRES (10) : " Shouldn't have. Why ? Because he was 

1 " Mfahant " which, for the sake of naturalness in the dialogue, 
we have rendered as " naughty ", conveys a sense of wickedness or 
violence which is not necessarily present in " vilain " [Trans.] 


stealing. What should he have done ? He should have 

JAC (n) : "He shouldn't have done it because the Ug boy 
would have nothing to eat. Should he have let himself be 
beaten? No. He should have taken his revenge. Told 
someone to help him to take his revenge, but not take his 

TRIP ^ (12), G. : "He wanted to take a revenge, but he 
shouldn't have. When people are unkind to us, we mustn't 
pay them back, we must tell our parents " 

It is easy to see what the attitude of these children is. 
Most of the little ones and a few of the older ones think 
that one should not take one's revenge, because there is a 
more legitimate as well as a more efficacious way of 
obtaining redress to call in the grown-up. With these 
children it is a question either of a petty calculation or of 
the predominance of the morality of authority over the 
morality of the relations between children. Telling tales 
(which constitutes an offence according to the latter code 
of morality) does not matter in the least, the great thing 
is to get fair treatment. For these children vengeance is 
wrong, but it is so at bottom because it has been forbidden. 
You must not render evil for evil, but you can have the 
fellow who wronged you punished. In addition to this, 
the little ones condemn the hero of the story because 
stealing is wrong, whatever may be the motive of the 
theft (moral realism). But with the older of the children 
quoted just now what predominates is not this complete 
submission or appeal made to adult justice it is the idea 
that there is not sufficient correspondence between the 
theft of the roll and the blows that have been received. 
Thus Jac, who is typical of this attitude, tells us quite 
clearly that the little boy ought to have returned the blows 
or got an older boy to return them, but not to have taken 
the roll. Justice, therefore, resides in reciprocity and not 
in mere brutal revenge. One should give back exactly 
what one has received, but not invent a sort of arbitrary 
punishment whose content bears no relation to the 
punishable act. These subjects are therefore not very far 


from approving of the hero of the story. At any rate they 
argue from the same reasons. 

Here are examples of those who approve of the little boy. 

MON (6J) : " He was quite right to have done it. Why ? 
Because the big boy was always beating him. Was it fair 
to hide his roll ? Yes. Was it right ? Yes." 

AUD : " He was quite right. Why ? Because his 
ldn't beat him. Was it fair to take his revenge ? 

brother sh 

. . . [Does not understand the word.] Was it naughty 

to do what the little boy did ? Not naughty." 

HEL (7^) : " He ought to have done it. Why? Because 
the big boy was always teasing him. Was it fair to do it ? 
Yes, it was fair. And was it right ?--. . . [Thinks it 
over] Yes, it is right." 

JAQ (7J), G. : " He was quite right. Why ? Because 
the big boy was always beating him. Was it fair ? Yes." 
But elsewhere Jaq answers : " Is it fair to be avenged ? 
Oh, no" In her eyes, therefore, the little boy's behaviour 
is not an act of vengeance but a punishment by reci- 

WID (8 ; 9) : " He should have done it because the big boy 
was always beating him. Was it fair ? Yes. Is it fair to 
take your revenge ? You mustn't take your revenge" 

CANT (9 ; 3) : " He should have done it. Why ? Be- 
cause he had been beaten. Was it fair to do it ? Yes. 
Was it right ? He shouldn't have hidden it. Why IHe 
need only have taken his revenge. How ? He ought to have 
kicked him." 

AG (10) : " He was quite right because the big boy was 
mean. Was it fair ? Yes, because big boys mustn't beat 
the little ones:' 

BACIN (n ; i), G. : "He was right to do it because he 
couldn't defend himself. Was it fair to do what he did ? 
Not very fair, because the big boy had been given the roll and 
the apple and then couldn't eat them. What would make it 
quite fair 1Hif him back: 1 Thus the little boy's theft is 
tolerated failing the correct punishment which would 
consist in giving back exactly what one had received. 

COLL (12 ; 8) : " In a way it is fair because there is 
nothing else he can do. In another way it is not fair to take 
his brother's roll" 

Thus we get one of two things. Either revenge is 
giving back exactly what you have received, and then it 


is fair (case of Cant), or else it is inventing in cold blood 
something unpleasant that will hurt the person who has 
hurt you and then it is unfair (case of Jaq, Wid, etc.). 
But all the children are agreed upon one point in the 
story : the little boy would have done best simply to give 
back the blows he had received ; but given the impossi- 
bility of such a procedure, he may be allowed to restore 
the balance of things by hiding the big boy's lunch. 

The second question should one hit back ? raises no 
difficulties, and the answers it calls forth are extremely 
simple. While they affirm in perfect good faith that one 
must not take one's revenge (in the special sense of being 
revenged in cold'blood) nor render evil for evil, the children 
maintain with a conviction that grows with their years 
that it is strictly fair to give back the blows one has 
received. Mile Rambert obtained the following statistics, 
taking girls and boys separately. 

" It is Give back Give back Give back 
naughty." the same. more. less. 

% % % 

(Girls . . 82 18 

A eb \Boys . . 50 37 "5 *2'5 

(G. ... 45 45 10 

7 IB. . . . 27 27 46 

JG. ... 25 42 8 25 

lB. ... 45 22 33 

JG. . . . 14 29 57 

9 IB. . . . 29 57 14 

fG. . , . 20 80 

10 IB. ... 8 54 31 7 

fG. ... - 33 - 67 

XI IB. ... 31 3^ 38 

/G. . . . 22 78 

I2 \B. ... 67 10 23 

This will show that in spite of inevitable irregularities 
of detail there exists, in girls as well as in boys, a tendency 
that increases with age to consider it legitimate to give 
back the blows one has received. Whereas more than half 
of the six-year-olds and a large proportion of those be- 
tween 7 and 8 still think that " It is naughty ", this 


answer almost completely disappears after the age of 9. 
But while this evolution is common to boys and girls, 
the former differ from the latter on the question of whether 
one should give back more, or less, or exactly the same 
as one has received. Boys, especially towards the age 
of 7-8, are inclined to give more, the desire for equality 
gaining the ascendent later towards 11-12. Girls, on 
the contrary, as soon as they have ceased in the majority 
of cases to think it " naughty " to hit back, are of opinion 
that one should give back less than one has received. 

Here, to begin with, are examples of those who think 
it " naughty " to hit back. 

JEA (6), G. : " If anyone hits you what do you do ? / 
tell teacher. Why do you not hit back ? Because it 
is naughty.' 9 

SAV (6) : "What do you do? I tell my mother. 
Do you hit back ? No, I'm afraid of being hurt. I go 
an<i tell teacher, so she'll punish him. Why must the 
teacher punish him? Because he's naughty (mdchant). 
If he has been naughty (m&chant), is it fair to punch 
him back again ? No, because that you would be [ = we 
would be] naughty (mechant)." 

BRA (6), G. : " What do you do ? I call my mother. 
Do you hit back ? No. Why do you call your mother ? 
Because he ought not to have given me a punch. Is it fair 
to hit back ? It's not fair, it's naughty (mechant)." 

Au (7 ; 9) : " I go and tell my daddy. And if he's not 
there ? / tell teacher. And if she is not there, do you 
hit back ? No. Why ? You get punished afterwards. 
Is it fair not to hit back ? Yes. People love us after- 
wards, and mummy and daddy are pleased." 

CHA (8) : " / tell teacher. I don't hit back. It's naughty 
(mechant) " 

NEN (9 ; 7), G. : " I don't hit back at all. I want to 
show her a good example. I am not horrid (mechant) to her." 

Here are examples of those who give back as many 
hits as they receive. 

PRA (6|) : " / don't let them hit me. How many punches 
do you give ? One for one. If he only gives me one, then 
I only give him one back. If he gives me two, I give him 
two. If he gives me three, I give him three. And ten ? I 


give him that back too. Is it fair to hit back ? Yes, it is 
fair. Why ? Because he had hit me too!' 

SCA (7$) : " / hit him back. I don't want it [the punch 
he gives me]. I give it to the other boy. Is that fair? 
Yes. Oh, no. I was wrong. You must never hit back. 
That's what my daddy said. But I'm like that. I won't let 
people punch me and kick me.' 3 Thus Sea knows Ms lesson, 
but he hits back all the same, and thinks it fair to do so. 

HEL (7$) : " I give him back two if he has given me two, 
six if he has given me six, four if he has given me four. 
Is it fair to hit back ? Yes, it is quite right." 

Die (8|), G. : "I defend myself. I hit back once for 
each hit. Why not oftener ? Because the other one would 
give me twice as much [and if I gave Mm back two for one, 
he would give me four]. Is it fair ? Yes. Three for three. 
You mustn't give in. You must defend yourself. Is that 
right ? Not very" [Die knows that it is not allowed.] 
(9 ; 7), G, : t( I hit her back again. How ma 

^ many 

hits do you give her ? As many as she has given me. 
Why not more ? So that they should add up the same. 
Is that right ? Yes." 

Pi (10) : "I give back one, and according how hard it is 
[the blow received], two. If you get five? Then I give 
five back. Why not more ? It would have hurt him more." 

ER (10 ; 2) had answered to question I that the little 
boy ought not to have stolen the big boy's roll and apple : 
" Why ? You mustn't take a revenge on people. Why ? 
Because it isn't nice." But when we ask Mm what he 
does when anyone gives him a punch, he answers, " / 
give him one back. And if you get two punches ? / give 
back two. You must never give more, otherwise the other 
boy gives you another again. Is it fair to hit back ? 
Yes. And to take a revenge ? Oh, no. To take a re- 
venge is not the same as punching back." 

HEN (n ; 2) : "I give him back a punch. If you are 
given two punches ? / give back two. If you are given 
three ? / give back three. W T hy not more ? Because I 
don't want to be worse than he is. I give him back his own 
hits. Is that fair ? No, because I ought to show myself 
better than he. Is taking your revenge the same thing as 
giving a punch ? It's not the same. To hit back is to give 
a punch. To take your revenge is mean." 

ELIS (ii), G. : " / hit back. Is it fair to do that ? 
[She hesitates.] Yes f it's fair. If anyone hits you once ? 
I hit back once. If I hit back twice, it isn't fair." 


Here are examples of boys who hit back oftener than 
they have been hit. 

JE (7) : " What do you do when anyone gives you 
a punch ? I give 'em back too. And if they give you 
three II give them back four. Is it fair to do that ? 
Yes: 9 

ET (10) : " // they give me one I give back two. If they 
give me two, I give back three" 

And of girls who give back less. 

BOE (8 ; 5), G. : " You ought to hit back. If they hit 
you three times? I hit back once. Why not three 
times ? That would be naughty (mechant). Is it fair to 
hit back ? No, you oughtn't to hit back/' 

BER (10), G. : "I hit him back less, because if I hit 
him back the same or more', he begins again. Is it right to 
hit back ? No, it's not right." 

It will be seen by these examples that the children 
who do not hit back (most of them are from among the 
younger ones), are primarily submissive children who 
rely upon the adult to protect them, and who are more 
anxious to respect or make others respect the orders that 
have been received than to establish justice and equality by 
methods appropriate to child society. As for the children 
who hit back, they are far more concerned with justice 
and equality than with revenge properly so called. The 
cases of Er and Hen are particularly clear. These children 
disapprove of cold-blooded revenge and petty scheming, 
but they uphold exact reciprocity from a sense of justice. 
Among those who give back more blows than they re- 
ceive there is, of course, a combative attitude which 
goes beyond mere equality ; but it is precisely this atti- 
tude which diminishes with age. 

Let us now turn to a question which will serve as a 
transition between retributive and distributive judgment 
between children why should one not cheat at games ? 
You ask the child what is his favourite game, and you 
tell him the story of a little boy who cheated (e.g. to 
change one's place more than necessary during a game 


of marbles). When the subject has stated that this is 
cheating, you ask him why one should not cheat at 
games. The answers fall under four heads : i It is 
naughty (forbidden, etc.). 2 It is contrary to the rules 
of the game. 3 It makes cooperation impossible 
(" You can't play any more "). 4 It is contrary to 

If we divide the children into two groups according 
to age, the first from 6-9 (it will be remembered that it 
was round about 9 that rules began to be stabilized), 
and the second from 10-12, we shall note the following 
changes as we pass from one group to another. Answers 
appealing to the authority of rules (whether of morality 
or games) i.e. answers of types I and II, drop from 70% 
to 32%, whereas answers of types III and IV, appealing, 
that is, to cooperation or to equality, rise from 30% to 
68%. These data can, moreover, be given in greater 
detail. First type answers (simply it is naughty, for- 
bidden, etc.) fall from 64% to 8% ; while those of the 
second type are 6% before, and 24% after 9. Third type 
answers (cooperation) rise from o% to 20%, and those of 
the fourth type (equality) from 30% to 48%. 

Such a result as this is easy to understand if our 
analysis of game rules be brought to mind. According 
to the little ones, in whom unilateral respect predominates 
and who identify a game rule with a moral rule, to 
cheat is " naughty ", like lying or using a rude word : 
it is forbidden by order and suppressed by punishment. 
Hence the abundance of type I answers before 8-9. This 
frequency might, of course, be explained by the difficulties 
which little children experience in analysing their thoughts, 
but there is, we believe, in addition to this, the moral 
element, to which we have just drawn attention. The 
older children, for whom rules have become a direct 
emanation of the autonomous group, condemn cheating 
for reasons which appeal to this very solidarity and to 
the resulting equalitarianism. 

Here are examples of the first type, 


DEM (6 ; 2): "It's naughty (Fr. vilain).Why?~ 
You must never cheat (Fr. frouiller). Why ? My [eldest] 
brother told, me." 

BRAIL (6) : " It's not fair. The others don't let us cheat 
(Fr. frouiller). They say, ' Go away ! ' " 

VAN (6|) : " Because you mustn't. It's naughty. 
Why ? Because it's very naughty. Why is it naughty ? 
Because you ought never to do it. Why not do it ? 
Because it's very bad. You never must. Why must you 
never ? Just because." 

GREM (7 ; 2) : " You mustn't. Why ? Because it's 
naughty. Why ? Because it's a bad thing to do." 

Gis (8), G. : " Because it isn't nice (Fr. /ofo*). Why ? 
Must never cheat. Why ? It's ugly. Why ? You must 
never cheat. It's very naughty." 

These arguments simply amount to this, that the 
thing is forbidden. The answers of the second type are 
not very different. 

ZUR (6 ; 6) : " It's not the game. Why ? Because he 
ought not to have done -it. Why ? Because the one who 
cheated (Fr. frouille) has spoilt the game. You can't have 
fun any more. Why ? It's naughty." 

CHRI (6 ; 10) : " You mustn't cheat. Why ? Because 
it isn't fair. Why ? Because it wouldn't finish the game. 
The game would be wrong." 

WAL (7}) : " It's not allowed because it's not the game." 

MARG (9) : " It's not doing it right. Why must people 
not cheat ? Because 'it's not the game. Why ? You 
mustn't. Why not ? It's not the game." 

And here are examples of the third type (cooperation). 

SCHA (7) : " People mustn't cheat. Otherwise you don't 
have them again [ = You don't play with the cheaters 
again]. You don't like them. Why ? Because you aren't 
good friends any longer. Why ? They get naughty (Fr. 

Go (7 ; 2) : " It upsets the game and makes the others 
angry. It mixes the game all up because he has made us 
angry. You can't play any more." 

BRU (9 ; 2) : "It spoils the game." 

Tis (10 ; i) : " You don't want to play any more. 
Why ? It isn't fair. Why ? If everyone did that no 
one II play any more." 


Wi (10) : " It's not doing it fight. It's deceiving the 
others. Why must you do things right in a game ? 
So as to be honest when we grow up! 9 [There is a boy who 
has understood how much more useful is a well-regulated 
game than a lesson in morals.] 

THEV (10), G. : " It's a bad action. Why IShe acted 
wrongly. She shouldn't have done that. And supposing 
she had lost ? It would have been better to lose "than to 
deceive. And supposing she had cheated and lost all the 
same ? She would have been punished [ipso facto]. It 
wasn't fair that she should have won. Why must we not 
cheat ? Because it's committing a lie. 3 ' 

PERO (10) : " You say to them, * // you are going to 
cheat we won't have you any more '. Why ? Because 
people who cheat are rotters (Fr. des sales types)." 

ZAC (n) : " It isn't nice. Why ? You can't have any 
nice fun. You call him liar." 

BOIL (12) : " If people cheat, it's not worth while playing. 19 

And finally some cases of the fourth type (equality). 

MER (9 ; 6) : " It's not fair. Why ? The others don't 
do it> so you mustn't do it either." 

THER (9 ; 7) : " It's not fair on the others." 

PER (n ; 9) : " It's not fair. You win what you haven't 
the right to [win]/* 

Gus (n) : " It's not fair. Why ? The others don't 
cheat, so it isn't fair." 

GAG (12 ; o) : " It would be unjust to the others" 

It is easy enough to see that between the answers 
appealing to cooperation and those that lay more stress 
on equality, there are innumerable intermediate stages. 
In the child as everywhere else solidarity and equali- 
tarianism are interdependent. In short, there would seem 
to be two fundamental types of answer the one appeal- 
ing to authority (types I and II), the other to cooperation 
(types III and IV). Between them, naturally, there 
are intermediate cases. Thus there is not complete 
heterogeneity between the answers given by Thev and 
those of type I. But broadly speaking, the two types are 
distinct, and the second gradually gains preponderance 
over the first. 


It may be worth while, in this connection, to recall 
the results of an enquiry made on lying (Chap. II, 4), 
and which also has a certain bearing upon the problem 
of equality between children. Is it as bad (vilain) * to lie 
to one's companions as to grown-ups, or is it different ? 
According to Mile Rambert's results, 81% of the subjects 
between 6 and 9 think it worse to lie to adults, while 
51% of those between 10 and 13 that it is equally bad to 
lie to children, and of these, 17% are even of opinion 
that it is worse to lie to a companion than to an adult. 

Let us now turn to the questions of distributive justice 
properly so-called, in the relations between children. 
We studied in this connection the two points that seemed 
to us most important equality between contemporaries, 
and the problem of differences in age. Here are two 
stories that were used for the analysis of the first of these 
two questions. 

STORY I. Some children are playing ball in a court- 
yard. When the ball goes out of bounds and rolls down 
the road one of the boys goes of his own free will to fetch 
it several times. After that he is the only one they ask 
to go and fetcfi it. What do you think of that ? 

STORY II. Some children are having their tea on the 
grass. They each have a roll that they have bought, 
and put it down beside them to eat after their brown 
bread. A dog comes up very softly behind one of the boys 
and snatches away his roll. What should be done ? 

We shall not require a lengthy analysis to sort out the 
answers : the children were unanimous in demanding 
equality. In the first story, it is not fair that it should 
always be the same child who works for the group, and 
in the second, each should give the victim enough to 
supply him with a share that will be equal with that of 
the others. We lay stress on these answers, simply be- 
cause in analogous stories, but where the desire for equality 

1 The word " vilain ", which we have throughout translated as 
" naughty ", can also mean " horrid ", and therefore ba.s a slightly 
different shade of meaning when used between children. [Trans.] 


found itself pitted against adult authority, the younger 
children very often, it will be remembered, put authority 
in the right ( 5). 
Here are some examples. 

WAL (6), Story I : " It isn't fair. Why ? Because 
another boy should go" Story II : " They must share. 
Why ? So as all to have the same." 

SCHMA (7), Story I : " It's not fair, because they should 
have asked the others, and each in turn." Now, in the story 
about the father who sent one boy on more messages 
than the other, Schma had answered, " It is fair, because 
the father said. . . ." 

As to Story II : " The others must share with him so 
that he should have a piece" Then we ask, just to see 
whether this desire for equality keeps authority at bay, 
" But if the mother doesn't want him to be given any 
more. She says he need only have stopped the dog taking 
the roll. Is it fair ? Yes. He need only have been 
careful. And if the mother had said nothing, what 
would have been fairest ? They ought to have shared" 

DELL (8), Story I : " It's not fair. They ought to have 
gone themselves." Story II : " They ought to have shared." 

ROB (9), Story I : " They should each have gone in turn" 
Story II : " They should have shared. The mother said 
they mustn't. That's not fair " 

FSCHA (10), Story I : " Another one ought to have gone" 
Story II : " They ought each to have shared a half with 
the boy who had none" 

We have already met with so many examples of the 
progressive development of equalitarianism ( 5), that 
we need press the point no further. 

But there remains the question as to what children 
think about differences of age. Should precedence be 
given to the seniors, or should the younger ones be 
favoured, or should all be treated equally ? We pre- 
sented the two following stories to our subjects : 

Story I. Two boys, a little one and a big one, once 
went for a long walk in the mountains. When lunch- 
time came they were very hungry and took their food 
out of their bags. But they found that there was not 


enough for both of them. What should have been done ? 
Give all the food to the big boy or to the little one, or the 
same to both ? 

Story II. Two boys were running races (or playing 
marbles, etc.). One was big, the other little. Should they 
both have started from the same place, or should the 
little one have started nearer ? 

The second question is complicated by the fact that 
the game is an organized one, and consequently regu- 
lated by tradition. The first, on the other hand, gave 
rise to a very interesting reaction. The younger children 
are in favour of equality or else, and chiefly, of precedence 
being given to the big boys out of respect for their age ; 
whereas the older children are in favour either of equality 
or else, and chiefly, of precedence being given to the little 
boys out of equity. 

Here are examples of the little ones 1 answers. 

JAN (7!) : " They should both have been given the same. 
They gave most to the little boy. Was it fair ? No. 
They should all have had the same. All half. Aren't 
little children more hungry ? Yes. If you had been the 
little boy what would you have done ? I'd have given 
less to myself and more to the big boys." 

NEV (7!) : " The big boys should have had most. Why ? 
Because they're bigger" 

FAL (7!) : " The big boy should have had most. Why ? 
Because he's the eldest. If you had been the little one, 
would you have given most to the big ones ? Yes. 
Ought they to have more, or is it that they want more ? 
They ought to have more" 

ROB (9) : " A little more to the big one. Why ? Be- 
cause he is the eldest. Who gets most hungry during 
walks, little boys or big ones ? Both the same. If you 
were out on a walk with a boy of twelve and there was 
only one piece of bread between you, what would you 
do ? I'd give him most. Would you think that was 
all right ? Yes } I'd want to give him most" 

Here are children in favour of equality. 

WAL (7) : " Each must be given the same. Why ? 
Another time they had five bars of chocolate. The 
little boy asked for three. Was it fair ? They ought to 


have had two and a half each. Supposing you are going 
for a walk with a boy, and you are the biggest, and keep 
most for yourself. Is it fair ? Not fair." 

Nuss (10) : " They ought to have gone shares. The 
little boy said, ' I'm the smallest so I ought to have most/ 
Is it fair ? Not fair. The big boy said he had right 
to most because he was the biggest. Is that fair? 
They ought each to have taken the same amount. You 
are ten. Suppose you are going for a walk with a boy 
of fifteen who gives you most of what there is, what 
would you think of it ? It would be nice of him. And 
fair ? It would be still more fair both to be given the same." 

And examples of equity. 

SCHMO (10) : " They should have given more to the little 
boy because he was smaller. They both ate the same. 
Was it fair? Not quite so fair' 9 

BRA (10) : " The same for everyone. They gave the 
little one most ; was it fair or not ? It was fair. Oh, 
no. The big boy kept most for himself, because he was 
the biggest. Was it fair ? It wasn't fair." 

As to the games, the answers differ according as the 
game is a race or marbles. Running races is relatively 
uncodified, and this freedom from rules allows for the 
little ones being favoured. In the game of marbles, the 
authority of the rules complicates the reactions. The 
little ones demand equality because it is the inviolable 
rule of the game, whereas the older boys are inclined to 
make exceptions in favour of the little ones. Here are 
two examples of the younger children's reactions. 

BRI (6). In running races : " The little boy must have 
a start because the big boy can run faster than the little one." 
But in marbles it must be " the same for both." Why ? 
Because [if both do not start from the same point and the 
little one is helped] God will make it happen that the big 
boy hits the marbles, and the little one can't" To make an 
exception is thus identified with cheating, which will be 
punished by divine justice. 

WAL (7) : In running races one must put " the smallest 
boy a little farther forward " t but at marbles " all at the 
coche [the starting-line]. Why ? You always start at the 


And one example of the reaction in older children. 

BRA (10) : In races the little ones must have a start. 
At marbles the same applies, " Because it is always done 
when they are two or three years younger." 

In conclusion, we find that the notions of justice 
and solidarity develop correlatively and as a function 
of the mental age of the child. In the course of this 
section, three sets of facts have appeared to us to be 
connected together. In the first place, reciprocity asserts 
itself with age. To hit back seems wrong to the little 
ones because it is forbidden by adult law, but it seems 
just to the older children, precisely because this mode 
of retributive justice functions independently of the 
adult and sets " punishment by reciprocity " above 
" expiatory punishment." In the second place, the desire 
for equality increases with age. Finally, certain features 
of solidarity, such as not cheating or not lying between 
children, develop concurrently with the above tendencies. 

enquiry to a close let us examine the answers given to a 
question which sums up all that we have been talking 
about. We asked the children, either at the end or at 
the beginning of our interrogatories, to give us them- 
selves examples of what they regarded as unfair. 1 

The answers we obtained were of four kinds : i Be- 
haviour that goes against commands received from the 
adult lying, stealing, breakages, etc. ; in a word, every- 
thing that is forbidden. 2 Behaviour that goes against 
the rules of a game. 3 Behaviour that goes against 
equality (inequality in punishment as in treatment). 
4 Acts of injustice connected with adult society (eco- 
nomic or political injustice). Now, statistically, the results 
show very clearly as functions of age : 

Forbidden Games Inequality Social Injustice 

6-8 . . 64% 9% 27% 

9-12 . . 7% 9% 73% 11% 

1 As a matter of fact this term is not understood by all, but it can 
always be replaced by " not fair " (Fr. pas juste). 


Here are examples of the identification of what is 
unfair with what is forbidden : 

AGE 6 : " A little girl who has broken a plate ", " to 
burst a balloon ", " children who make a noise with their 
feet during prayers ", " telling lies " t " something not 
true ", " it's not fair to steal ", etc. 

AGE 7 : " Fighting ", " disobeying ", "fighting about 
nothing " t " crying for nothing ", " playing pranks ", etc. 

AGE 8 : " Fighting each other ", " idling lies ", " steal- 
ing ", etc. 

Here are examples of inequalities : 

AGE 6 : " Giving a big cake to one and a little one to 
another" " One piece of chocolate to one and two to another." 

AGE 7 : " A mother who gives more to a little girl who 
isn't nice." lt Beating a friend who has done nothing to 

AGE 8 : " Someone who gave two tubes [to two brothers] 
and one was bigger than the other " [taken from experience, 
this !] " Two twin sisters who were not given the same 
number of cherries " [also experienced]. 

AGE 9 : " The mother gives a [bigger] piece of bread 
to someone else." (t The mother gives a lovely dog to one 
sister and not to the other." " A worse punishment for one 
than for the other." 

AGE 10 : " When you both do the same work and don't 
get the same reward." " Two children both do what they 
are told, and one gets more than the other" " To scold one 
child and not the other if they have both disobeyed" 

AGE ii : " Two children who steal cherries : only one 
is punished because his teeth are black" "A strong man 
beating a weak one" " A master who likes one boy better 
than another, and gives him better marks" 

AGE 12 : "A referee who takes sides" 

And some examples of social injustice. 

AGE 12 : "A mistress preferring a pupil because he is 
stronger, or cleverer, or better dressed" 

11 Often people like to choose rich friends rather than poor 
friends who would be nicer" 

" A mother who won't allow her children to play with 
children who are less well dressed" 

" Children who leave a little girl out of their games, who 
is not so well dressed as they are." 


These obviously spontaneous remarks, taken together 
with the rest of our enquiry, allow us to conclude, in so 
far as one can talk of stages in the moral life, the existence 
of three great periods in the development of the sense 
of justice in the child. One period, lasting up to the age 
of 7-8, during which justice is subordinated to adult 
authority ; a period contained approximately between 
8-ii| and which is that of progressive equalitarianism ; 
and finally a period which sets in towards 11-12, and 
during which purely equalitarian justice is tempered by 
considerations of equity. 

The first is characterized by the non-differentiation 
of the notions of just and unjust from those of duty and 
disobedience : whatever conforms to the dictates of the 
adult authority is just. As a matter of fact even at this 
stage the child already looks upon some kinds of treat- 
ment as unjust, those, namely, in which the adult does 
not cany out the rules he has himself laid down for children 
(e.g. punishing for a fault that has not been committed, 
forbidding what has previously been allowed, etc.). But 
if the adult sticks to his own rules, everything he pre- 
scribes is just. In the domain of retributive justice, 
every punishment is accepted as perfectly legitimate, as 
necessary, and even as constituting the essence of morality : 
if lying were not punished, one would be allowed to tell 
lies, etc. In the stories where we have brought retributive 
justice into conflict with equality, the child belonging 
to this stage sets the necessity for punishment above 
equality of any sort. In the choice of punishments, 
expiation takes precedence over punishment by recipro- 
city, the very principle of the latter type of punishment 
not being exactly understood by the child. In the domain 
of immanent justice, more than three-quarters of the 
subjects under 8 believe in an automatic justice which 
emanates from physical nature and inanimate objects. 
If obedience and equality are brought into conflict, the 
child is always in favour of obedience : authority takes 
precedence over justice. Finally, in the domain of justice 


between children, the need for equality is already felt, 
but is yielded to only where it cannot possibly come into 
conflict with authority. For instance, the act of hitting 
back, which is regarded by the child of 10 as one of ele- 
mentary justice, is considered " naughty " by the children 
of 6 and 7, though, of course, they are always doing it 
in practice. (It will be remembered that the heterono- 
mous rule, whatever may be the respect in which it is 
held mentally, is not necessarily observed in real life.) 
On the other hand, even in the relations between children, 
the authority of older ones will outweigh equality. In 
short, we may say that throughout this period, during 
which unilateral respect is stronger than mutual respect, 
the conception of justice can only develop on certain 
points, those, namely, where cooperation begins to make 
itself felt independently of constraint. On all other 
points, what is just is confused with what is imposed 
by law, and law is completely heteronomous and imposed 
by the adult. 

The second period does not appear on the plane of 
reflection and moral judgment until about the age of 
7 or 8. But it is obvious that this comes slightly later 
than what happens with regard to practice. This period 
may be defined by the progressive development of au- 
tonomy and the priority of equality over authority. In 
the domain of retributive justice, the idea of expiatory 
punishment is no longer accepted with the same docility 
as before, and the only punishments accepted as really 
legitimate are those based upon reciprocity. Belief in 
immanent justice is perceptibly on the decrease and 
moral action is sought for its own sake, independently 
of reward or punishment. In matters of distributive 
justice, equality rules supreme. In conflicts between 
punishment and equality, equality outweighs every other 
consideration. The same holds good a fortiori of conflicts 
with authority. Finally, in the relations between children, 
equalitarianism obtains progressively with increasing 


Towards 11-12 we see a new attitude emerge, wMch 
may be said to be characterized by the feeling of equity, 
and which is nothing but a development of equalitari- 
anism in the direction of relativity. Instead of looking 
for equality in identity, the child no longer thinks of the 
equal rights of individuals except in relation to the par- 
ticular situation of each. In the domain of retributive 
justice this comes to the same thing as not applying the 
same punishment to all, but taking into account the 
attenuating circumstances of some. In the domain of 
distributive justice it means no longer thinking of a law 
as identical for all but taking account of the personal 
circumstances of each (favouring the younger ones, etc.). 
Far from leading to privileges, such an attitude tends to 
make equality more effectual than it was before. 

Even if this evolution does not consist of general 
stages, but simply of phases characterizing certain limited 
processes, we have said enough to try to elucidate 
now the psychological origins of the idea of justice 
and the conditions of its development. With this in 
view, let us distinguish retributive from distributive 
justice, for the two go together only when reduced to 
their fundamental elements, and let us begin with dis- 
tributive judgment, whose fate in the course of mental 
development seems to indicate that it is the most funda- 
mental form of justice itself. 

Distributive justice can be reduced to the ideas of 
equality or equity. From the point of view of episte- 
mology such notions cannot but be regarded as a priori, 
if by a priori we mean, not of course an innate idea, but 
a norm, towards which reason cannot help but tend as 
it is gradually refined and purified. For reciprocity im- 
poses itself on practical reason as logical principles impose 
themselves morally on theoretical reason. But from the 
psychological point of view, which is that of what is, 
not of what should be, an a priori norm has no exist- 
ence except as a form of equilibrium. It constitutes 
the ideal equilibrium towards which the phenomena 


tend, and the whole question is still to know why, 
the facts being what they are, their form of equilibrium 
is such and no other. This last problem, which is of a 
causal order, must not be confused with the first, which 
can be solved only by abstract reflection. The two will 
coincide only when inind and reality become coextensive. 
In the meantime let us confine ourselves to psychological 
analysis, it being understood that the experimental 
explanation of the notion of reciprocity can in no way 
contradict its a priori aspect. 

From this point of view it cannot be denied that the 
idea of equality or of distributive justice possesses in- 
dividual or biological roots which are necessary but not 
sufficient conditions for its development. One can observe 
in the child at a very early stage two reactions which will 
play a very important part in this particular elaboration. 
Jealousy, to begin with, appears extremely early in babies : 
infants of 8 to 12 months often give signs of violent rage 
when they see another child seated on their mother's 
knees, or when a toy is taken from them and given to 
another child. On the other hand, one can observe in 
conjunction with imitation and the ensuing sympathy, 
altruistic reactions and a tendency to share, which are 
of equally early date. An infant of 12 months will hand 
his toys over to another child, and so on. But it goes 
without saying that equalitarianism can never be regarded 
as a sort of instinct or spontaneous product of the in- 
dividual mind. The reactions we have just alluded to 
lead to a capricious alternation of egoism and sympathy. 
It is true, of course, that jealousy prevents other people 
from taking advantage of us, and the need to communi- 
cate prevents the self from taking advantage of others. 
But for true equality and a genuine desire for reciprocity 
there must be a collective rule which is the sui generis 
product of life lived in common. There must be born 
of the actions and reactions of individuals upon each 
other the consciousness of a necessary equilibrium binding 
upon and limiting both "alter" and "ego". And this 


ideal equilibrium, dimly felt on the occasion of every 
quarrel and every peace-making, naturally presupposes 
a long reciprocal education of the children by each other. 

But between the primitive individual reactions, which 
give the need for justice a chance of showing itself, and the 
full possession of the idea of equality, our enquiry shows 
the existence of a long interval in time. For it is not 
until about 10-12, at the age where, as we saw elsewhere, 
children's societies attain to the maximum of organization 
and codification of rules, that justice really frees herself 
from all her adventitious trappings. Here, as before, we 
must therefore distinguish constraint from cooperation, 
and our problem will then be to determine whether it is 
unilateral respect, the source of constraint, or mutual 
respect, the source of cooperation, that is the preponder- 
ating factor in the evolution of equaHtarian justice, 

Now on this point the results of our analysis seem to 
leave no room for doubt. Authority as such cannot be 
the source of justice, because the development of justice 
presupposes autonomy. This does not mean, of course, 
that the adult plays no part in the development of justice, 
even of the distributive kind. In so far as he practises 
reciprocity with the child and preaches by example 
rather than by precept, he exercises here, as always, 
an enormous influence. But the most direct effect of 
adult ascendancy is, as M. Bovet has shown, the feeling 
of duty, and there is a sort of contradiction between 
the submission demanded by duty and the complete 
autonomy required by the development of justice. For, 
resting as it does on equality and reciprocity, justice can 
only come into being by free consent., Adult authority 
even if it acts in conformity with justice, has therefore 
the effect of weakening what constitutes the essence of 
justice. Hence those reactions which we observed among 
the smaller children, who confused what was just with 
what was law, law being whatever is prescribed by adult 
authority. Justice is identified with formulated rules 
as indeed it is in the opinion of a great many adults, of 


all, namely, who have not succeeded in setting autonomy 
of conscience above social prejudice and the written 

Thus adult authority, although perhaps it constitutes 
a necessary moment in the moral evolution of the child, 
is not in itself sufficient to create a sense of justice. This 
can develop only through the progress made by co- 
operation and mutual respect cooperation between 
children to begin with, and then between child and adult 
as the child approaches adolescence and comes, secretly 
at least, to consider himself as the adult's equal. 

In support of these hypotheses, one is struck by the 
extent to which, in child as well as in adult society, the 
progress of equalitarianism goes hand in hand with that 
of " organic " solidarity, i.e. with the results of co- 
operation. For if we compare the societies formed by 
children of 5-7 with those formed at the age of 10-12, 
we can observe four interdependent transformations. In 
the first place, while the little ones' society constitutes 
an amorphous and unorganized whole, in which all the 
individuals are alike, that of the older children achieves 
an organic unity, with laws and regulations, and often 
even a division of social work (leaders, referees, etc.). 
In the second place, there exists between the older children 
a far stronger moral solidarity than among the younger 
ones. The little ones are simultaneously egocentric and 
impersonal, yielding to every suggestion that comes along 
and to every current of imitation. In their case the group 
feeling is a sort of communion of submission to seniors and 
to the dictates of adults. Older children, on the contrary, 
ban lies among themselves, cheating, and everything that 
compromises solidarity. The group feeling is therefore 
more direct and more consciously cultivated. In the third 
place, personality develops in the measure that discussion 
and the interchange of ideas replace the simple mutual 
imitation of the younger children. In the fourth place, 
the sense of equality is, as we have just seen, far stronger 
in the older than in the younger children, the latter 


being primarily under the domination of authority. 
Thus the bond between equaEtarianism and solidarity 
is a universal psychological phenomenon, and not, as 
might appear to be the case in adult society, dependent 
only upon political factors. With children as with 
adults, there exist two psychological types of social 
equilibrium a type based on the constraint of age, 
which excludes both equality and " organic " solidarity, 
but which canalizes individual egocentrism without 
excluding it, and a type based on cooperation and resting 
on equality and solidarity. 

Let us pass on to retributive justice. In contrast to 
the principles of distributive justice, there does not seem 
to be in the ideas of retribution or punishment any pro- 
perly rational or a priori element. For while the idea 
of equality gains in value as intellectual development 
proceeds, the idea of punishment seems actually to lose 
ground. To put things more precisely, we must, as we 
have already done, distinguish two separate elements 
in the idea of retribution. On the one hand there are the 
notions of expiation and reward, which seems to con- 
stitute what is most specific about the idea of punishment, 
and on the other, there are the ideas of " putting things 
right " or making reparation, as well as the measures 
which aim at restoring the bond of solidarity broken by 
the offending act. These last ideas, which we have 
grouped under the title of " punishment by reciprocity " t 
seem to draw only on the conceptions of equality and 
reciprocity. It is the former set of ideas that tends to be 
eliminated when the morality of heteronomy and authority 
is superseded by the morality of autonomy. The second set 
are of far more enduring stuff, precisely because they are 
based upon something more than the idea of punishment. 

Whatever may be said of this evolution of values, it is 
possible here, as in connection with distributive justice, 
to assign three sources to the three chief aspects of re- 
tribution. As we saw above ( i) certain individual 
reactions condition the appearance of retribution ; adult 


constraint explains the formation of the idea of expiation, 
and cooperation accounts for the eventual fate of the idea 
of punishment. 

It cannot be denied that the idea of punishment has 
psycho-biological roots. Blow calls for blow and gentle- 
ness moves us to gentleness. The instinctive reactions 
of defence and sympathy thus bring about a sort of 
elementary reciprocity which is the soil that retribution 
demands for its growth. But this soil is naturally not 
enough in itself, and the individual factors cannot of 
themselves transcend the stage of impulsive vengeance 
without finding themselves subject at least implicitly 
to the system of regulated and codified sanctions implied 
in retributive justice. 

Things change with the intervention of the adult. 
Very early in life, even before the infant can speak, its 
conduct is constantly being subjected to approval or 
censure. According to circumstances people are pleased 
with baby and smile at it, or else frown and leave it 
to cry, and the very inflections in the voices of those 
that surround it are alone sufficient to constitute an 
incessant retribution. During the years that follow, the 
child is watched over continuously, everything he does 
and says is controlled, gives rise to encouragement or 
reproof, and the vast majority of adults still look upon 
punishment, corporal or otherwise, as perfectly legitimate. 
It is obviously these -reactions on the part of the adult, 
due generally to fatigue or impatience, but often, too, 
coldly thought out on his part, it is obviously these adult 
reactions, we repeat, that are the psychological starting- 
point of the idea of expiatory punishment. If the child 
felt nothing but fear or mistrust, as may happen in extreme 
cases, this would simply lead to open war. But as the 
child loves his parents and feels for their actions that 
respect which M. Bovet has so ably analysed, punish- 
ment appears to him as morally obligatory and necessarily 
connected with the act that provoked it. Disobedience 
the principle of all " sin ''is a breach of the normal 


relations between parent and child ; some reparation is 
therefore necessary, and since parents display their 
" righteous anger " by the various reactions that take 
the form of punishments, to accept these punishments 
constitutes the most natural form of reparation. The 
pain inflicted thus seems to re-establish the relations that 
had momentarily been interrupted, and in this way the 
idea of expiation becomes incorporated in the values of 
the morality of authority. In our view, therefore, this 
" primitive " and materialistic conception of expiatory 
punishment is not imposed as such by the adult upon 
the child, and it was perhaps never invented by a psycho- 
logically adult mind; but it is the inevitable product 
of punishment as refracted in the mystically realistic 
mentality of the child. 

If, then, there is such close solidarity between the idea 
of punishment and unilateral respect plus the morality 
of authority, it follows that all progress in cooperation 
and mutual respect will be such as to gradually eliminate 
the idea of expiation from the idea of punishment, and 
to reduce the latter to a simple act of reparation, or a 
simple measure of reciprocity. And this is actually what 
we believe we have observed in the child. As respect 
for adult punishment gradually grows less, certain types 
of conduct develop which one cannot but class under the 
heading of retributive justice. We saw an example of 
this in the judgments made by our subjects on the topic 
of " hitting back " ; the child feels more and more that 
it is fair that he should defend himself and to give back 
the blows he receives. This is retribution without doubt, 
but the idea of expiation seems not to play the slightest 
part in these judgments. It is entirely a matter of re- 
ciprocity. So-and-so takes upon himself the right to give 
me a punch, he therefore gives me the right to do the 
same to him. Similarly, the cheat gains a certain advant- 
age by the fact of cheating ; it is therefore legitimate 
to restore equality by turning him out of the game or by 
taking back the marbles he has won. 


It may be objected that such a morality will not take 
one very far, since the best adult consciences ask for 
something more than the practice of mere reciprocity. 
Charity and the forgiving of injuries done to one are, in 
the eyes of many, far greater things than sheer equality. 
In this connection, moralists have often laid stress on the 
conflict between justice and love, since justice often pre- 
scribes what is reproved by love and vice versa. But in 
our view, it is precisely this concern with reciprocity 
which leads one beyond the rather short-sighted justice 
of those children who give back the mathematical equiva- 
lent of the blows they have received. Like all spiritual 
realities which are the result, not of external constraint 
but of autonomous development, reciprocity has two 
aspects : reciprocity as a fact, and reciprocity as an 
ideal, as something which ought to be. The child begins 
by simply practising reciprocity, in itself not so easy a 
thing as one might think. Then, once he has grown 
accustomed to this form of equilibrium in his actions, 
his behaviour is altered from within, its form reacting, 
as it were, upon its content. What is regarded as just 
is no longer merely reciprocal action, but primarily be- 
haviour that admits of indefinitely sustained reciprocity. 
The motto " Do as you would be done by ", thus comes to 
replace the conception of cnide equality. The child sets 
forgiveness above revenge, not out of weakness, but 
because "there is no end" to revenge (a boy of id). 
Just as in logic, we can see a sort of reaction of the form 
of the proposition upon its content when the principle 
of contradiction leads to a simplification and purification 
of the initial definitions, so in ethics, reciprocity implies 
a purification of the deeper trend of conduct, guiding it 
by gradual stages to universality itself. Without leaving 
the sphere of reciprocity, generosity the characteristic 
of our third stage allies itself to justice pure and simple, 
and between the more refined forms of justice, such as 
equity and love properly so called, there is no longer 
any real conflict. 


In conclusion, then, we find in the domain of justice, 
as in the other two domains already dealt with, that 
opposition of two moralities to which we have so often 
drawn the reader's attention. The ethics of authority, 
which is that of duty and obedience, leads, in the domain 
of justice, to the confusion of what is just with the content 
of established law and to the .acceptance of expiatory 
punishment. The ethics of mutual respect, which is 
that of good (as opposed to duty), and of autonomy, 
leads, in the domain of justice, to the development of 
equality, which is the idea at the bottom of distributive 
justice and of reciprocity. Solidarity between equals 
appears once more as the source of a whole set of com- 
plementary and coherent moral ideas which characterize 
the rational mentality. The question may, of course, 
be raised whether such realities could ever develop without 
a preliminary stage, during which the child's conscience 
is moulded by his unilateral respect for the adult. As 
this cannot be put to the test, by experiment, it is idle to 
argue the point. But what is certain is that the moral 
equilibrium achieved by the complementary conceptions 
of heteronomous duty and of punishment properly so 
called, is an unstable equilibrium, owing to the fact that 
it does not allow the personality to grow and expand to 
its full extent. As the child grows up, the subjection of 
his conscience to the mind of the adult seems to him 
less legitimate, and except in cases of arrested moral 
development, caused either by decisive inner submission 
(those adults who remain children all their lives), or by 
sustained revolt, unilateral respect tends of itself to grow, 
into mutual respect and to the state of cooperation which 
constitutes the normal equilibrium. It is obvious that 
since in our modern societies the common morality 
which regulates the relations of adults to each other is 
that of cooperation, the development of child morality 
will be accelerated by the examples that surround it. 
Actually, however, this is more probably a phenomenon 
of convergence than one simply of social pressure. For 


if Iranian societies have evolved from heteronomy to 
autonomy, and from gerontocratic theocracy in all its 
forms to equalitarian democracy, it may very well be 
that the phenomena of social condensation so well de- 
scribed by Durkheim have been favourable primarily to 
the emancipation of one generation from another, and 
have thus rendered possible in children and adolescents 
the development we have outlined above. 

But having reached the point where the problems of 
sociology meet those of genetic psychology, we are faced 
with a question of too great moment to allow us to rest 
content with these indications, and we must now compare 
our results with the fundamental theses of sociology and 
psychology concerning the empirical nature of the moral 



WHETHER we wish it or not, the questions we have had to 
discuss in connection with child morality take us to the 
very heart of the problems studied by contemporary 
sociology and social psychology. Society, according to 
Durkheim's followers, is the only source of morality. If 
this is so, then there is no discipline in a better position 
to discover it than child psychology. Every form of 
sociology will in fact inevitably lead to a system of 
pedagogy, as readers of Durkheim's excellent book on 
L'Education Morale will have had occasion to note. It is 
for this reason that we intend, in spite of the difficulties 
attending an attempt of this kind, to examine some of 
the more significant of the sociological and ethico-psycho- 
logical theories of the day, and to compare them with our 
own findings. We shall abstain from any discussion that 
would be too general in character, and shall confine our- 
selves to those spheres in which the theories of social 
psychology in vogue have a direct bearing upon the child. 
With this aim in view, the following points seem to us 
to deserve examination. The ideas of Durkheim and 
Fauconnet on responsibility, which gave rise to Durkheim's 
writings on punishment in schools, Durkheim's theory of 
authority as the source of the moral life of the child, the 
theories of M. Baldwin, and above all those of M. Bovet 
on the genesis of the moral sentiment, and finally certain 
educational ideas concerning autonomy of conscience in 
the child. 

But in order to prevent ary misunderstanding, we must 
point out from the first that in these discussions on 



educational matters proper, It is as psychologists and not 
as educationalists that we approach the subject. Educa- 
tional facts are facts of social psychology, perhaps, indeed, 
the most important of their number, and while we do not 
wish to establish a system of education on the results of 
our enquiry, we cannot abstain from asking, for example, 
whether a given authoritarian procedure is, as Durkheim 
maintains, really necessary to the constitution of the moral 
life. This question is one of pure, as well as of applied, 
psychology. Whether a cure recommended by some doctors 
will kill the patient or cure him is a matter that interests 
the physiologist as well as the doctor, and it is as an 
experimenter, not as a practitioner, that we wish to speak 
of the subject of education. 

RESPONSIBILITY. 1 M. Paul Fauconnet in Ms excellent 
book on La Responsabiliti has developed Durkheim's ideas 
on retributive justice and penal law in a striking and 
original manner. He ends by showing that the earliest 
and, in his view, the purest form of responsibility is no 
other than the objective responsibility of which we found 
so many examples in connection with children. No better 
theme could be better found, therefore, to assist us in the 
critical examination of what probably constitutes the 
quintessence of Durkheim's teaching, the idea, namely, 
that society is always one and the same, and that its 
permanent features are such as to ensure the existence and 
invariability of moral values. 

Responsibility is, according to M. Fauconnet, the 
" quality belonging to those who must ... in virtue of a 
rule be chosen as the passive subjects of a punishment " 
(p. n) ; to be responsible is to be " justly punishable " 
(p. 7). Now the comparative study of different societies 
about which we have sufficient information allows us to 
establish a sort of evolutionary law which dominates the 
whole history of responsibility. Starting with far richer 

1 P. Fauconnet, La Responsabilitt, Etude de Sociologie, Alcan. 1920. 


and more comprehensive forms, responsibility has only 
gradually shrunk to its present dimensions. To begin 
with, in contemporary civilized communities, responsible 
subjects consist exclusively of adults who are alive and in 
their right minds. In older or non-civilized societies, as 
in the Middle Ages, and frequently in more recent times, 
responsible subjects included children, insane persons (even 
recognized as such), the dead, animals and above all collec- 
tive groups as such. The study of situations generative of 
responsibility leads to the same conclusion. In our com- 
munities, intention (or other psychological features, such 
as negligence, forgetfulness, etc.) is a necessary condition 
of responsibility. Ethically, the intention is everything. 
Legally, there must be a corpus delicti, but there is no 
offence without intent, imprudence or negligence. Now, 
" as we trace our way back through the history of penal 
law, we gradually come to purely objective responsi- 
bility " (p. 105). In other words, in primitive ethics or in 
archaic law, responsibility is ascribed even to involuntary 
acts, accidents, acts committed without either negligence 
or imprudence. In short, primitive responsibility is above 
all objective and communicable, ours is subjective and 
strictly individual. 

What, then, is responsibility ? In order to solve this 
problem, M. Fauconnet, true to the spirit of Durkheim, 
seeks to explain the phenomena, not by establishing the 
laws governing their making or growth but by empha- 
sizing the elements that are invariable and common to 
every stage. Note should be taken of this fundamental 
point of method. Thus philosophical explanations of 
responsibility are to be put aside as ignoring the early 
forms of the phenomenon in question. Evolutionary 
doctrines, like Westermark's and others, have this dis- 
advantage that they reduce the early forms to moral or 
intellectual aberrations, as though our present conceptions 
constituted the supreme norm or necessary outcome of all 
that preceded them. If, on the contrary, we ascribe the 
same value to all cases brought to light by the compara- 


tive method, we shall see that even if the punishment does 
not maintain an unequivocal and well-defined relation to 
the victim, it is always clearly determined in relation to 
the crime. In other words, crime has been punished in all 
times and in all places, and if the punishment does not 
fall directly upon the head of the culprit, it will always 
reach someone or other. A crime taken as a material act 
independent of the motives involved is a sort of centre of 
infection which must be destroyed, together with every- 
thing around it, both far and near. "The penalty is 
directed against the crime. It is only because it cannot 
reach the crime that it is deflected on to a substitute 
for the crime " (p. 234). 

This is where the theory of crime elaborated by Durk- 
heim comes in. Every society consists primarily in a 
collection of beliefs and feelings forming a whole which 
must be defended. The kernel of these beliefs is the 
feeling of the sacred, the source of all morality and re- 
ligion. Whatever offends against these powerful and well- 
defined feelings is crime, and all crime is sacrilege. A 
crime that breaks down the social bond takes on, by the 
mere fact of doing so, a mystical significance. It is a 
source of impurity and contamination, and its repercussions, 
visible and invisible, are incalculable. It must therefore 
be suppressed, its disastrous consequences must be sup- 
pressed, and things must be put right. Punishment is the 
mystical procedure that will effect this restitution. 

Consequently it matters very little on whom the punish- 
ments fall. The great thing is that they should be inflicted 
and that they should be proportionate to the crime. Thus 
there is an " institution of responsibility ". Moreover, it 
is easy to understand how the choice of the responsible 
subject comes to be made. The process takes place in 
virtue of a mechanism of transference which obeys the 
usual laws of psychological transference. First, there is an 
affective transference : the emotions aroused by the crime 
are carried over to everything that touches it from near 
or far. Then there is a judgment : the community 


decides that a given individual is responsible, and this 
judgment is dominated by relations of contiguity and 
resemblance. It follows, naturally, that the culprit himself, 
when he can be found, is held to represent the maximum of 
relationship with the crime. But failing this, anything 
that touches the crime must be punished. Thus responsi- 
bility descends from outside upon the culprit or any of his 
substitutes, and transforms them into scapegoats or instru- 
ments of social purification. Thus responsibility has a very 
definite function : " to make the realization of the penalty 
possible by letting it play its useful part " (p. 297). Now 
this rdle is essentially moral : " It is only on the condition 
that punishments exist that the existence of morality itself 
is ensured : punishment, and consequently responsibility 
have therefore a share in the value of morality " (p. 300). 

But there remains a problem. How is it, if this is the 
essence of responsibility, that this " institution " should 
have evolved to such a point that we are no longer able, 
at first sight, to make head or tail of its primitive forms ? 
How has responsibility come to shrink in this way .and to 
be directed on none but the one intentional, adult, and 
normal culprit ? Why has responsibility individualized 
and spiritualized itself in this fashion ? 

Far from constituting the necessary culmination in an 
inner transformation of responsibility, its contemporary 
form results, on the contrary, according to Fauconnet, from 
a gradual weakening of primitive values due to the action of 
antagonistic factors. The general cause of this evolutionary 
process would therefore seem to be external to responsi- 
bility itself it is pity and humanitarianism. For though 
the community may be outraged by crime, antagonistic 
feelings will appear at the moment of punishment, and, as 
Ihering has said, the story of penalties is the story of 
constant abolition. This is why responsibility shows a 
constant tendency to shrink. 

In the beginning, the community will punish anyone, 
and the individual is only a means to an end. In our day, 
we punish almost against our wills, and the culprit is 


given every chance of defending himself and of escaping 

This gives rise to two fundamental consequences. In 
the first place, " we may say that in the course of its 
evolution, responsibility becomes individualized. Collec- 
tive and communicable in elementary communities, it 
is, in principle, strictly personal in the most civilized 
societies." Only in Theology, that is to say, in the most 
conservative of our institutions, does the idea of Original 
Sin keep alive the idea of collective responsibility ; Adam's 
fall leaves upon the whole of humanity a stain that calls 
for expiation. In law and in ethics such ideas would 
revolt us. But our purely individual responsibility is 
nothing but a bastard form of the real thing. " We are 
generally taught that responsibility is individual by nature, 
communicable by accident. The history of responsi- 
bility is interpreted as a progress, and true responsibility, 
strictly personal in character, is supposed to have been 
achieved in the course of evolution. We are led, however, 
to present the facts under a totally different light. The 
expanding and contagious character of responsibility has 
seemed to us to be its fundamental feature. The indi- 
vidualization of responsibility is the result, on the contrary, 
of a limiting and attenuating process. Far from purifying 
it and perfecting it, the forces that individualize responsi- 
bility are antagonistic to its nature. Strictly personal 
responsibility is like the last positive value of a responsi- 
bility that is tending towards zero-point. From this point 
of view, the evolution of responsibility appears as a 
regression. What one takes to be perfect responsibility 
is responsibility whittled down to vanishing point" 

(PP. 343~4)- 
Hence a second aspect of the evolution of responsibility 

its spiritualization. Primitive responsibility is above all 
objective. A crime is "first and foremost a material 
event. And the bond uniting the crime to whoever is 
responsible for it is nearly always, to begin with, a 
material bond " (p. 345). But, " in the eyes of our con- 


temporaries, responsibility arises in the conscience of the 
man who is responsible, on the occasion of a spiritual act, 
by reason of a psychological relation between the con- 
science and the act. These features contrast line for line 
with the objectivity described above " (pp. 345-6)- The 
cause of this phenomenon of spiritualization is that 
society, which begins by being external to men's minds, 
" becomes more and more immanent to the individual. 
More and more of him becomes socialized with time, and 
the contribution of social life is added to and modifies 
what is of psycho-organic origin. The spiritualization of 
moral or religious ideas expresses this veritable penetra- 
tion of the social into the individual " (p. 367). In short, 
if responsibility takes account of intentions only, it does 
so in virtue of the same process as that by which it has 
become individualized. "As social life becomes indi- 
vidualized it becomes more interiorized " (p. 351). So 
that our moral consciousness is nothing but the interiorized 
residuum of the collective consciousness. But this is not 
a gain. " Like the individualization of responsibility, its 
spiritualization in the course of history appears therefore 
as an immense impoverishment, a perpetual abolition. 
Subjective responsibility, far from being, as is generally 
thought, responsibility par excellence, is only an atrophied 
form of responsibility " (p. 350). 

But one must not be too absolute. There are still in 
our societies traces of collective and objective responsi- 
bility, and M. Fauconnet claims to find examples of it in 
the present trend of criminal law in Italy. The primitive 
form has therefore not been completely killed by the 
existing form. " So true is it that the former is still a 
living branch on the common stock from which the latter 
has gradually become detached" (p. 377). From the 
educational point of view, it will therefore be natural to 
come with Durkheim to the conclusion of the necessity for 
a systematic method of school punishments as the only 
means of reviving in men's minds the permanent source of 
all responsibility. 


For ourselves we know of no thesis so well suited as this 
doctrine of Durkheim's and Fauconnet's to throw light on 
the problems raised by the affirmation of the moral unity 
of society. For M. Fauconnet has accomplished a tour de 
force. He has tried to explain by the permanence be- 
longing to the conditions of the collective consciousness a 
phenomenon of which he, more successfully than anyone 
else, has traced the evolution down history. Nor do we 
wish to dispute M. Fauconnet's subtle analysis and 
masterly classification of the facts he has collected. The 
kinship between the phenomena of objective responsibility 
that can be observed in the child and the phenomena that 
characterize primitive social constraint furnishes, on the 
contrary, an excellent confirmation of the hypothesis that 
the initial externality of social relations is bound to bring 
with it a certain moral realism. What seems to us more 
questionable is the general interpretation he gives of re- 
sponsibility, of punishment, and of the relations holding 
between moral facts and society regarded as a fixed and 
unchanging whole. 

The great lesson of comparative sociology is that there 
exist at least two types of responsibility one objective 
and communicable, the other subjective and individual, 
and that social evolution has gradually caused the second 
to predominate. Having established this, two solutions 
were equally conceivable : to define responsibility by the 
direction it took in history, by its vector, or to define it 
by its constant structural elements. 
, On the strength of Durkheim's methodological postulates, 
M. Fauconnet has chosen the second solution. He tells us 
at once that what interests him in the history of responsi- 
bility is not so much the transformations as the invariables. 
And what leads him to this choice is the deep and funda- 
mental unity of social facts. " However diverse civiliza- 
tions may be, there is such a thing as civilization " (p. 20). 
But it may be questioned whether such a method is 
satisfactory beyond a given point. What would we think 
of a psychologist who, in his explanation of causality and 


number, put magical and mystical views of causality on 
the same level as Einstein's treatment of causality or as 
the theory of complex numbers ? If you look only for 
the elements that are common to all stages and eliminate 
all considerations of direction, you will achieve nothing 
but a flimsy and static residue (what, after all, do the 
elements common to all the different stages of causality 
amount to ?), or else you will lay all the stress as we fear 
M. Fauconnet has done on the primitive forms. Now 
it is, no doubt, indispensable to be acquainted with the 
primitive forms of the facts one wishes to analyse, but 
their subsequent evolution furnishes data of at least equal 
importance upon the conditions of their genesis. To put 
the matter more clearly, we should try both to determine 
the direction of the successive stages, and to establish the 
invariables common to all these stages. But what is 
invariable is not a given structural feature otherwise the 
primitive form could always be erected into the " true " 
form but only the function. As for the structure, it 
varies indefinitely in so far as its variations are subjected 
to its function, and the laws of evolution controlling such 
variations are more instructive than the features peculiar 
to any particular stage. 

It is neither with the laws of the evolution of responsi- 
bility that M. Fauconnet deals, nor with the function that 
is common to its different stages. He tells us clearly why 
it is not the laws of its evolution. And neither is it the 
common function, taken independently of the structure of 
the phenomenon, since nowadays our responsibility is 
no longer either ^communicable or objective, and since 
M. Fauconnet considers these two features as constitutive 
of " true " responsibility. 

Now the psychological data of child morality suggest 
to us an interpretation of responsibility which, while it 
takes full account of the valuable results obtained by 
M. Fauconnet, seems to us to fulfil the double claim 
of invariability or functional continuity and of directed 
structural evolution. For we have recognized the exist- 


ence of two moralities in the child, that of constraint and 
that of cooperation. The morality of constraint is that of 
duty pure and simple and of heteronomy. The child 
accepts from the adult a certain number of commands to 
which it must submit whatever the circumstances may be. 
Right is what conforms with these commands ; wrong is 
what fails to do so ; the intention plays a very small part 
in this conception, and the responsibility is entirely 
objective. But, first parallel with this morality, and then 
in contrast to it, there is gradually developed a morality 
of cooperation, whose guiding principle is solidarity and 
which puts the primary emphasis on autonomy of con- 
science, on intentionality, and consequently on subjective 
responsibility. Now it should be noted that while the 
ethics of mutual respect is, from the point of view of values, 
opposed to that of unilateral respect, the former is never- 
theless the natural outcome of the latter from the point of 
view of what causes this evolution. In so far as the child 
tends towards manhood, his relations with the adult tend 
towards equality. The unilateral respect belonging to 
constraint is not a stable system, and the equilibrium 
towards which it tends is no other than mutual respect. 
It cannot, therefore, be maintained with regard to the 
child that the final predominance of subjective over 
objective responsibility is the outcome of antagonistic 
forces in relation to responsibility in general. Rather is 
it in virtue of a sort of inner logic that the more 
evolved follow upon the more primitive forms, though 
in structure the former differ qualitatively from the 

Why, then, in an extremely schematic way, should the 
same thing not hold good of society ? It is no mere 
metaphor to say that a relation can be established between 
the individual's obedience to collective imperatives and 
the child's obedience to adults in general. In both cases 
the human being submits to certain commands because he 
respects his elders. Society is nothing but a series (or 
rather many intersecting series) of generations, each exer- 


cising pressure upon the one which follows it, and Auguste 
Comte was right in pointing to this action of one genera- 
tion upon the other as the most important phenomenon 
of sociology. Now when we think of the part played by 
gerontocracy in primitive communities, when we think 
of the decreasing power of the family in the course of 
social evolution, and of all the social features that charac- 
terize modem civilizations, we cannot help seeing in the 
history of societies a sort of gradual emancipation of 
individuals ; in other words, a levelling up of the different 
generations in relation to each other. As Durkheim him- 
self has pointed out, one cannot explain the passage from 
the forced conformity of " segmented " societies to the 
organic solidarity of differentiated societies without in- 
voking the diminished supervision of the group over the 
individual as a fundamental psychological factor. The 
" denser " the community, the sooner will the adolescent 
escape from the direct constraint of his relations and, 
coming under a number of fresh influences, acquire his 
spiritual independence by comparing them with one 
another. The more complex the society, the more au- 
tonomous is the personality and the more important are 
the relations of cooperation between equal individuals. 
Now, if cooperation follows as naturally as this upon 
constraint, and consequently the morality of mutual re- 
spect upon that of authority, there seems no reason why 
subjective responsibility should be looked upon as a 
bastard form of " primitive " responsibility. The most 
one could say would be that rational mentality was a 
bastard form of prescientific or " prelogical " mental- 
ity, and that independent morality was an attenuated 
form of elementary religions. It seems to us, on the 
contrary, that if the phenomena of education may legiti- 
mately be regarded as exhibiting in embryo the principles 
of social phenomena, then subjective responsibility is the 
normal outcome of objective responsibility, in so far, at least, 
as the constraint of conformity gives place to a coopera- 
tion based upon social differentiation and upon individ- 


ualism. So that it is not only because pity and humani- 
tarianism exercise a check upon penalties that objective 
and communicable responsibility is weakened. It is 
because the ideas of involuntary crime and even of 
expiatory punishment lose all meaning in the morality of 
autonomy. Conceptions of offence and punishment evolve 
at the same time as those of duty, of right and wrong, 
and of distributive justice. The two types of morality 
distinguished by M. Fauconnet seem to us, therefore, to 
be connected with two general attitudes, with the two 
moralities we have grown accustomed to, and the evolu- 
tion from one to the other is not contingent but goes of 
a piece with the psycho-sociological transformation that 
characterizes the passage from the theocratic conformity 
of so-called " primitive " societies to the equalitarian soli- 
darity of modern times. The heteronomy that goes with 
constraint begets objective responsibility, just as the 
autonomy that goes with mutual respect and cooperation 
begets subjective responsibility. For the morality of 
cooperation, the germs of which are present in any society, 
is stifled by the constraint prevalent in conformist so- 
cieties. But the social differentiation and " organic " 
solidarity peculiar to civilized communities allow it to 
expand, and thus explain the transformation of responsi- 

Let us examine M. Fauconnet's arguments more closely 
and make our choice between the two interpretations. 
Although he brings out very clearly the difference between 
our conceptions and those of primitive times, M. Fauconnet 
tries to soften the contrast by looking for signs of objective 
responsibility and even for promising revivals of archaic 
notions in our own times. Thus in speaking of the Italian 
school of jurisprudence, which regards the criminal .as an 
irresponsible degenerate, society itself being responsible 
for his condition, M. Fauconnet declares : " Only by 
returning to its own source can responsibility be renewed 
and sustained " (p. 344). Thus the responsibility of the 
community in the case of the delinquent drunkard is com- 



pared to that of the clan in the case of an individual 
violation of the taboo. But at this point, we fear that a 
confusion has crept in. If elementary communities look 
upon responsibility as collective, they do so in order to 
extend, as it were, the surface of application of the 
penalty, and in order the better to chastise and thus 
increase the mystical efficacy of the expiation. But in 
declaring society responsible for the crime of one of its 
members, what the modern mind seeks to do is, on the 
contrary, to diminish the latter's responsibility and to 
soften his punishment. In regarding a drunkard as the 
victim of society we seek to remove all intentionality and 
precisely all element of guilt from his misdeeds. Of course, 
traces of collective responsibility are still to be found 
among us, as when in war a whole nation is made re- 
sponsible for the faults of its rulers, or in religion where 
the whole of humanity is condemned to perdition because 
of the sins of its first progenitors. These certainly are 
instances of collective responsibility. But such judgments 
are not generally made by the normal collective conscious- 
ness and should be regarded as residua, or regressions, 
rather than as a " renovation " of responsibility. 

Another argument used by M. Fauconnet is that in 
penal law there must be a corpus delicti. But this is a 
circumstance of a practical nature rather than an ideal 
principle to be adhered to by the collective consciousness. 
If judges could really " try the reins and the heart ", 
if, without risk of error, with the same or even greater 
certainty than they use in establishing the proof of simple 
matters of fact, they could lay bare the intention behind 
the act, then it is undoubtedly the guilty intention itself 
that would incur punishment. It is therefore for lack of 
infallibility and not because of any positive principles of 
objective responsibility that justice confines its activities 
to material acts, in which restricted domain it already 
encounters difficulties enough. 

In short, modern responsibility tends to be entirely 
individual and entirely subjective. But one must go 


further than this. Between the primitive conception of 
responsibility and certain existing ideas on the subject 
there is a difference, not only of degree, but of kind. In 
the case of the first, punishment is the moral conception 
par excellence. And as the primitive notions are all 
realistic, the idea of punishment, for lack of differentiation 
between the psychical and the physical, has, to begin 
with, a meaning that is as much magical as mystical. 
The crime being a centre of physical and psychical 
infection, its punishment will have effects that are both 
material and spiritual. Except in regard to certain theo- 
logical conceptions, these ideas have become completely 
foreign to our way of thinking. At the same time, the idea 
that a good action deserves its reward and an evil one its 
punishment, is one that is still very widespread. This 
conception, which combines very easily with that of 
subjective responsibility, we willingly allow to be a product 
of primitive, objective, and quasi-materialistic notions of 
responsibility. But it will be admitted that a great many 
of our contemporaries have freed themselves of this idea. 
To the minds of many, not only, needless to say, should 
good be sought for its own sake, but punishment of any 
kind is immoral. If, with M. Fauconnet, we identify 
responsibility with punishment, it will be only in reference 
to those chosen few that we shall talk of an " attenuation 
of responsibility". But the whole point about these 
people is that they dissociate responsibility from punish- 
ment, and reject punishment as energetically as they 
cultivate the idea of responsibility. Of course we must 
distinguish here between the legal and the moral point of 
view. From a purely legal point of view, punishment is 
perhaps necessary for the defence of society, though 
modern writers on the subject also tend to place the idea of 
social re-education and re-adaptation above that of expia- 
tion. But from the moral point of view, there is always 
something ambiguous about the idea of punishment, and 
the least we can say of it is that it renders autonomy of 
conscience impossible. M. Guyau's brilliant defence of 


tbis theory will be remembered, and if it lacked a certain 
depth, the same charge can hardly be laid to one of the 
most substantial reactions ever made against the Kantian 
Ethics, which, like so many, only went half the way to true 
autonomy we mean the ethical system of F. Rauh. 

In a word, inner responsibility, which goes with autonomy 
of conscience and arises from the relations of cooperation, 
is not simply derived from the form of responsibility that 
is bound up with the idea of expiatory punishment, and 
consequently with constraint and heteronomy : a new 
type of moral attitude has replaced a superannuated 
attitude, and the continuity of the two events in time 
must not be allowed to hide the difference of their nature. 

In conclusion and this is what we want to establish, 
for it will be invaluable to us in what follows social 
constraint and cooperation lead to results that do not 
admit of comparison. Social constraint and by this we 
mean any. social relation into which there enters an ele- 
ment of authority and which is not, like cooperation, the 
result of an interchange between equal individuals has 
on the individual results that are analogous to those 
exercised by adult constraint on the mind of the child. 
The two phenomena, moreover, are really one and the 
same thing, and the adult who is under the dominion of 
unilateral respect for the " Elders " and for tradition is 
really behaving like a child. It may even be maintained 
that the realism of primitive conceptions of crime and 
punishment is, in certain respects, an infantile reaction. 
To primitive man, the moral and the physical universe are 
one and the same thing, and a rule is both a law of nature 
and a principle of conduct. For this reason, crime 
threatens the very existence of the universe and must be 
mystically set at naught by a suitable expiation. But 
this idea of a law that is both physical and moral is the 
very core of the child's conception of the world ; for 
under the effect of adult constraint the child cannot con- 
ceive the laws of the physical universe except in the guise 
of a certain obedience rendered by things to rules. As to 


ideas of punishment and expiation, how could they have 
become so widespread in the adult community if men had 
not all first been children, and if the child had not been 
from the very beginning of his mental development re- 
spectful towards the decisions of the adult who reprimands 
him and punishes him ? Under the effects of social 
differentiation and cooperation, on the contrary, the indi- 
vidual is less and less dominated by the cult of the past 
and by the forced conformity which accompanies it. He 
then becomes really adult, and the infantile traits that 
mark the conformist spirit make place for the features 
that are the outcome of cooperation. Thus autonomy of 
conscience takes the place of heteronomy, whence we have 
the transformations studied by M. Fauconnet in the domain 
of responsibility, and the criticisms which come to be raised 
against the very idea of expiatory punishment ; thus the 
purely interior responsibility which blames itself for not 
reaching a certain ideal follows upon the responsibility 
born of the reactions of the group. It is true that this 
inner responsibility remains a social phenomenon ; unless 
individuals cooperated, conscience would be ignorant of 
right and of the sense of guilt. But this is a phenomenon 
of different order from that of the facts of constraint, 
though it constitutes in a fashion the form of equilibrium 
towards which the whole history of responsibility tends. 

I. INTRODUCTION. Durkheim's ethical teaching, which 
strikes so sincere a note and is imbued with such a deeply 
scientific spirit, certainly raises the gravest question we 
have to face in our interpretation of the facts of child 
psychology. Where we see a struggle between two 
moralities, and between two types of social relations, 
Durkheim affirms the unity of all moral and social facts. 
True, no one has realized better than he the deep socio- 
logical foundation for the struggle between independent 
and transcendent morality, but while we believe the first to 
have been prepared for by the solidarity amongst children 


themselves and the second to have proceeded from the 
constraint of the adult over the child, Durkheim regards 
all morality as imposed "by the group upon the individual 
and by the adult upon the child. Consequently, from the 
pedagogic point of view, whereas we would be inclined to 
see in the " Activity" School, in " self-government ", and 
in the autonomy of the child the only form of education 
likely to produce a rational morality, Durkheim upholds 
a system of education which is based on the traditional 
model and relies on methods that are fundamentally 
those of authority, in spite of all the tempering features 
he introduced into it in order to allow for inner liberty 
of conscience. 

Before examining L'Education Morale let us devote a 
little thought to the problem of Durkheim's moral soci- 
ology in general. For Durkheim's ideas on the child can 
only be understood in terms of his general sociology, and 
as his ideas on children resemble so closely those of 
common sense and current pedagogy we shall have to 
examine them very closely if we wish to uphold the 
soundness of our own theories on the subject. 

In his book on the division of labour, La Division du 
Travail Social, the least dogmatic and, theoretically, the 
most suggestive of his works, Durkheim has shown greater 
caution than he did later on with regard to the unity of 
social facts and consequently the identity of moral facts 
with each other. There are two great types of society : 
on the one hand, there are the conformist communities 
whose solidarity is segmented or mechanical ; and on the 
other, there are the communities differentiated from within 
by division of labour and possessing organic solidarity. 
The first are exclusive of inner freedom and of ^person- 
ality, the second mark the growth and expansion of 
individual dignity. Now, social differentiation is a recent 
phenomenon, as yet hardly defined and apt to over- 
throw our social habits and moral rules. By causing 
a break-down in traditional conformity, differentiation 
applies a check to the theological symbols connected with 


conformity, and since in communities of tlie first type 
morality depends upon religion and the external forms 
of religion, our first duty will be actually to create 
a new morality for ourselves. Our equilibrium is 
threatened. We need an inner equivalent of the outer 
solidarity that belongs to conformity. What we lose 
in the material constraint of traditional institutions we 
must gain in inner morality, in a personal cultivation of 

It seemed, then, according to this presentation of the 
subject, that social pressure appeared under two different 
manifestations : constraint properly so called in conformist 
societies, and inner obligation in cooperation. It seemed, 
at any rate, as though moral feelings would have to present 
themselves under almost opposite forms according as they 
belonged to the heteronomy of an obligatory conformity 
or to the autonomy belonging to personality in differen- 
tiated and organic societies. The tendency of Durkheim's 
later works, however, was to reduce constraint and co- 
operation to unity, and above all to fuse into a single 
explanation the analyses he gave of the different aspects 
of morality. 

In the case of constraint, Durkheim has stretched his 
definition of the word to cover all social phenomena what- 
soever. Whether he is talking of the inward attraction felt 
by the individual for universal human ideals, or of the 
coercion exercised by public opinion or police, it is all 
constraint. The " externality " of the social phenomenon 
gives rise to the same generalizations. Logical and moral 
principles are external to the individual in the sense that 
the individual mind would not have been able to work 
them out unaided. But so are verbal signs, mystical 
symbols or economic values, in the sense that it is not in 
the individual's power to alter them at will. In a word, 
society is always one and the same, and the differences 
between cooperation and obligatory conformity are more 
a matter of degree than of quality. 

Morality is treated in the same way. Morality is 


born of religion, since obligatory acts were first sanc- 
tioned in so far as they proceeded from the idea of 
the sacred. Just as the sacred is what inspires both a 
respectful fear and a feeling of attraction, so do moral 
conceptions present two irreducible, but inseparable 
aspects obligation and duty on the one hand, and 
on the other, the sense of the good or of a desirable 

Again, just as the sacred gives rise to ritual prohibitions 
and positive commands, so does morality forbid and 
compel without giving reasons. The Categorical Impera- 
tive is thus a direct emanation from social constraint. 
The object of morality and the source of respect can be 
only society itself, as distinct from individuals and 
superior to them. 

On the main thesis of Durkheim's doctrine, i.e. the 
explanation of morality by social life, and the interpre- 
tation of its changes in terms of the varying structure 
of society, we can naturally only agree with his socio- 
logical school, because of the results of our enquiry. We 
found that the purely individual elements of morality 
could be traced either to the feeling of respect felt by the 
younger for the older children, which explained the genesis 
of conscience and duty, or to the feelings of sympathy felt 
by the child for those around him, which made coopera- 
tion possible. Instinctive tendencies, together with others 
more or less directly connected with them, are thus a 
necessary but not a sufficient condition for the formation 
of morality. Morality presupposes the existence of rules 
which transcend the individual, and these rules could only 
develop through contact with other people. . Thus the 
fundamental conceptions of childish morality consist of 
those imposed by the adult and of those born of col- 
laboration between children themselves. In both cases, 
that is to say, whether the child's moral judgments are 
heteronomous or autonomous, accepted under pressure or 
worked out in freedom, this morality is social, and on this 
point Durkheim is unquestionably right. 


But the matter does not end here, and it remains to be 
seen whether the unity of all social facts postulated by Durk- 
heim is not such as to rob morality of its most fundamental 
and most specific characteristic its normative autonomy. 
The danger of the sociological explanation and Durkheim 
was the first to notice it is that it may compromise 
morality by identifying it with reasons of state, with 
accepted opinions, or with collective conservatism ; in a 
word, with everything that the greatest reformers have 
attacked in the name of conscience. And we do not think 
that the solution given by Durkheim to this crucial 
problem is likely to remove our doubts. We ask the 
reader's forbearance in laying so much stress on this much 
discussed problem. But it is one that in spite of any 
contrary appearance is so important in child psychology 
Durkheim's pedagogy proves this if nothing else that 
we must not be afraid of appearing to utter truisms in 
our effort to remove all possible misunderstanding. 

Durkheim seems to have remained undecided on this 
essential point. In some passages he seems to attribute 
moral value to prevalent opinion by the mere fact that 
it is prevalent. " The affinity between habit and moral 
practice is even such that any collective habit almost 
inevitably presents a certain moral character. Once a 
form of behaviour has become habitual in a group, any 
departure from it provokes an impulse of disapproval very 
closely akin to that called forth by moral faults properly 
so called. Such habits command, in a measure, the 
special respect that is paid to moral practices. While all 
collective habits are not moral, all moral practices are 
undoubtedly collective habits. Consequently, whoever is 
refractory to all that is habit may very easily prove so to 
morality ' ' (Education Morale, p. 3 1) . But in other places 
and the passages we shall now quote seem to us to express 
Durkheim's own ideal far better than those in which he 
defends " collective habit " Durkheim seems to say the 
exact opposite. " It has been objected to this view that it 
subjects the mind to the prevailing moral opinion. But 


this is not so, for the society which morality tells us to 
desire is not society as it appears to itself but society as it 
really is or tends to be. And it may happen that society's 
own consciousness of itself, acquired in and through 
opinion, may be an inadequate reflection of the under- 
lying reality. Public opinion may be full of anachronisms 
which make it lag behind the actual state of society " 
(Sociologie et Philosophic, p. 51). And most important of 
all, in answer to a criticism by M. Parodi, Durkheim 
did not hesitate to declare before the " Soci<*t fran$aise 
de Philosophic " that according to his system . of 
Ethics it was the consciences of great individuals that 
were in the right when they came into conflict with 
public opinion. It may be granted, for instance, that 
public opinion in Athens condemned Socrates, but it 
was he who was in the right and not the Athenians, 
because "Socrates expressed more faithfully than his 
judges the morality that suited the society of his time " 

Clearly, one must make one's choice between these two 
solutions. But this is just what Durkheim could not 
admit, for the choice would at the same time settle the 
problem of constraint and cooperation. For, either society 
is one, and all social processes, including cooperation, are 
to be assimilated to pure constraint alone, in which case 
right is bound to be determined by public opinion and 
traditional use ; or else, a distinction must be made 
between actual and ideal society, that is to say, between 
" opinion " and society such as it is " really tending to 
be ", in which case one is inevitably led to distinguish 
between constraint and cooperation more fundamentally 
than Durkheim wished to do, so as to place moral values 
above reasons of state. 

For how, we would ask, is it possible to distinguish 
between society as it is and society as it is tending to 
become ? Under a r&gime of compulsory conformity, i.e. 
of more or less unadulterated social constraint, such a 
distinction is unthinkable. A given set of beliefs and 


practices are handed to the individual, and morality con- 
sists in preserving them as such. Involuntary alterations 
may, of course, occur in the established usage, and the 
breach may become appreciable between the new and the 
traditional usage. But the ideal is at one's back, not 
before one's eyes. For what constraint imposes is an 
already organized system of rules and opinions ; you can 
take this system or leave it, but any form of argument or 
personal interpretation is irreconcilable with conformity. 
In differentiated communities like ours, on the contrary, 
new relations between individuals become conceivable, 
germs of which, indeed, were already to be found in 
primitive communities, particularly in matters of technical 
labour. In the midst of the network of groupings which 
constitute our present society, individuals agree, not so 
much to preserve a set of dogmas and rites, as to apply 
a " method " or set of methods. What one affirms is 
verified by the others; what one has done is tried out 
and tested by the others. The essence of experimental 
behaviour whether scientific, technical, or moral con- 
sists, not in a common belief, but in rules of mutual control. 
Everyone is free to bring in innovations, but only in so far 
as he succeeds in making himself understood by others 
and in understanding them. This cooperation, which, it 
must be admitted, is in many spheres still far from pre- 
vailing over social constraint, though it constitutes the 
ideal of democratic communities, is the only thing that 
allows for the distinction between what is and what ought 
to be. For the word "method" implies that certain 
provisional truths have been established, but above all 
that there is more to discover and that progress de- 
pends on compliance with certain norms. In this way 
the characteristics of the ideal are safeguarded. The 
essence of social constraint and of external authority, on 
the contrary, is to identify what is with what ought to be, 
the ideal state of things being thus conceived as already 
Durkheim's refusal to admit any other difference than 


one of degree between constraint and cooperation must be 
due to the fact that the two processes are connected by 
an endless series of intermediate links. As he showed in 
his first work, it is to the extent that a community grows 
in volume and in density that division of labour, indi- 
vidualism and cooperation triumph over compulsory con- 
formity. The hypothesis may therefore be advanced that 
in all communities the social totality exercises constraint 
upon the individuals, and that this constraint takes .the 
form of cooperative solidarity in our civilization, whereas 
it remains limited to a compulsory conformity in primitive 
communities. Only this factual continuity, which we in 
no way wish to deny, does not exclude a qualitative 
contrast in the results. One may conceive of cooperation 
as constituting the ideal form of equilibrium towards which 
society tends when compulsory conformity comes to 
break down. As this ideal is approached, the character 
of social life comes to differ qualitatively from its primitive 
state, although the movement between the two forms is 

Let us approach the matter as psychologists in order 
to see more clearly its relationship to the two moralities 
of the child. Durkheim, except on rare occasions, has in 
mind only society as made up of adults. Everything 
happens, in his books, as though the group as a total 
synthesis simply exerted its pressure upon all individuals 
independently of their age. Now let us imagine a com- 
munity of which the members had always been each 
other's contemporaries, and had lived their lives without 
experiencing either the constraint of the generations pre- 
ceding them, or the education of the generations follow- 
ing them. However many centuries we allow these 
individuals to have lived, and however collective we 
picture their psychology, the chances are a thousand to 
one against their ever reproducing the sociology to which 
we have been accustomed in the communities described 
as " primitive ". These people would gradually discover a 
language, a logic, a science, an ethic, and a metaphysic of 


their own, but allowing for groping experiments, for errors, 
and for momentary collective deviations, one cannot 
imagine anything that would lead them to accept a social 
constraint as firmly crystallized as those of the com- 
munities in which compulsory conformity holds its sway. 
For it seems to us obvious that all the elementary social 
phenomena would be radically different from what they 
are if communities had never been foimed except by indi- 
viduals of the same age and ignorant of the pressure of 
one generation upon the other. The younger child feels 
respect for the older, and for its parents and the more 
simply constituted the society in which he lives, the more 
durable a part does this unilateral respect pay in the life 
of an individual, as the respect for age and elders in the 
more primitive communities seems to show. Without this 
unilateral respect one simply does not see how the ethics 
and the logic peculiar to social constraint and conformity 
could ever have come into being. In the moral sphere, it 
may very well be that such facts as ritual obligations and 
prohibitions, moral realism and objective responsibility 
would not exist without the respect which the child feels 
for the adult. But one can go a step farther and sur- 
mise that the outstanding features of " primitive men- 
taEty " can be explained by a conjunction of the childish 
mentality with the effects of the constraint exercised by 
one generation upon the other. Primitive mentality would 
therefore be due to social constraint being refracted 
through the childish mind. 1 In our civilization, on the 
contrary, with its foundation of cooperation and individual 
differentiation, the egocentric mentality of the child hardly 
enters into fundamental social phenomena except in the 
case of those which constitute "survivals", as Durkheim 
has called them, or which are " behind-hand in relation to 

1 In saying this, we do not, of course, wish to return to the pre- 
sociological phase of psychology, but simply to show that within the 
framework established by sociology there is everything to be said for 
reinstating psychological analysis. There is at the present day far more 
parallelism than antagonism between sociological studies and psycho- 
logical research. 


the existing state of society/* The development of the 
sciences, of industry, the economic division of labour, 
rational morality, and democratic ideas, all seem to 
be so many conquests standing in no relation to the 
constraint of the generations upon one another and aris- 
ing directly from a cooperation that is independent of 

In short, social constraint may be regarded from the 
psychological point of view as arising partly from the 
constraint exercised by the adult upon the child, and as 
exercising in consequence of this a " consolidating " effect 
upon child mentality. This assumption gains force if, as 
we showed in the course of our enquiries, social constraint 
does not really suffice to " socialize " the child but accen- 
tuates its egocentrism. Cooperation, on the other hand, 
seems to be essentially the social relation which tends 
to eliminate infantile phenomena. And this is enough to 
show that cooperation, while constituting the ideal form of 
equilibrium towards which constraint approaches in so far 
as social condensation liberates the younger generations, 
leads eventually to results that are qualitatively the 
opposite of constraint. It shows finally that if, with 
Durkheim, we wish to distinguish between opinion and 
reason, between the observance of custom and that 
of moral norms, we must at the same time make 
a vigorous distinction between a social process such 
as constraint, which simply consecrates the existing 
order of things, and a social process such as coopera- 
tion, which essentially imposes a method and thus 
allows for the emancipation of what ought to be from 
what is. 

We can now turn to the examination of the really 
ethical part of Durkheim's sociology. Two parts should 
be distinguished here : the theory of duty or of moral 
obligation, and the theory of good or autonomy of 
conscience. With regard to duty we cannot but sub- 
scribe to Durkheim's theory, at any rate from the 
point of view of static sociology. It seems to us 


incontestable not only that the totality of duties in a 
given society is bound up with the structure of this 
society, but also that the very form of duty (the feeling 
of obligation) is connected with the constraint exercised 
by the society on the individuals. From the genetic point 
of view, on the other hand, it can be maintained that the 
pressure of an adult upon a child is sufficient to give rise 
to the sense of duty in the latter's mind and to do so 
independently of the pressure to which the adult is 
subjected on the part of the whole society. But we do not 
regard these facts as contradictory of Durkheim's teach- 
ing, whatever Durkheim himself might have thought about 
it. It is therefore unnecessary to press the point any 
further, the more so as we shall deal with it again in 
connection with M. Bovet's theory. 

The theory of the morally good, on the other hand, 
seems to us to persent difficulties which arise from those 
very relations between constraint and cooperation of 
which we have just spoken. Very rightly, Durkheim 
begins by contrasting good and duty, and pronounces 
these two aspects of the moral life so different as to be 
irreducible. And yet soon after this, Durkheim almost 
identifies them in so far as they seem to him inseparable, 
and for him every moral act partakes both of the obligation 
belonging to duty and of the desirability which character- 
izes what is good. And these two notions are given to us 
as having a common origin : good and duty both come 
from the feeling of the " sacred ", the sacred being at once 
imperative and desirable, just as is society itself, of which 
the sacred is only the reflection. 

Now this common origin is just what seems to us 
questionable. True, there is no feeling of duty without 
desirability, and therefore without a certain feeling of the 
good. But the reason for this is clear. Unilateral respect, 
which is at the back of all consciousness of duty, consists 
of a sui generis mixture of fear and love, which implies 
in consequence an element of desirability. But the con- 
verse is not true. Actions can be good, yet devoid of 


any element of obligation. 1 There are even individuals, 
as Durkheim himself remarked, to whom goodness matters 
far more than duty, and the converse is also true. 

Durkheim, it is true, explains these individual differ- 
ences in such a way as to fit them into his theory. " Each 
individual/' he says, " expresses the common morality in 
his own way ; each understands it, envisages it from a 
different angle ; perhaps no one mind is completely 
adequate to the morality of its own time " (Social, d 
Phil., p. 56). In consequence, then, there would seem to 
be a " common morality " existing as such in society and 
regarded by the various individuals each from his own 
point of view. This would explain why in reality the 
good and duty are indissociable, each individual being free 
to stress one or other of these two aspects of the moral 
life, which would thus be rendered mutually irreducible. 

But we shall have to examine more closely those re- 
lations between common morality and individuals ; for 
the whole of the part played by society in the genesis of 
moral ideas is here the point at issue. When a number of 
travellers are climbing a mountain or exploring a tract of 
country, one can say that their individual points of view 
are always inadequate, because they cannot see everything 
that there is to be seen, or because they cannot see it 
simultaneously. If " common morality " constitutes an 
object that is thus external to the individuals, it goes 
without saying that the separate minds will always be 
inadequate, and that their respective viewpoints will be 
reconcilable only in the absolute of such a realism. But 
another solution is possible, and one which Durkheim does 
not seem to have envisaged, namely that " common 
morality " does not consist in a " thing " given to indi- 
viduals from without, but in a sum of relations between 
individuals. Common morality would thus be defined by 
the system of laws of perspective enabling one to pass 

1 According to Durkheim (Sociologie et Philosophic, p. 65) no moral 
acts are " purely desirable, for they always require effort." But effort 
is not necessarily the same thing as obligation. 


from one point of view to the other, and allowing in 
consequence the making of a map or objective repre- 
sentation of the mountain or country. In this case each 
individual perspective could be different from the others 
and yet at the same time adequate and in no danger of 
jeopardizing the coherence of the whole. It is true that this 
whole is extremely complex and that the individual con- 
sciences are also perhaps inadequate in the first sense of the 
term. But the point is that the second sense should also 
have its share of truth, as we believe to be really the case. 
For while duty constitutes a collection of commands more or 
less identical in each case, the good demands in addition a 
certain margin of personal elaboration and autonomy. 
Which surely means that far from being the result of 
constraint, it can be explained only by cooperation. 

In this way we are led to examine the curious argumen- 
tation by means of which Durkheim defines the object of 
what is morally good. This ob j ect cannot be the individual, 
since egoism is rejected by every kind of ethics. There- 
fore, " if each individual taken separately is incapable of 
giving any moral value to conduct, i.e. has in himself no 
moral value, neither does the numerical sum of such 
individuals possess any value " (Social, et Philos., p. 73). 
Thus, " now that we have eliminated the individual 
subject, there remains no possible objective for moral 
activity than the sui generis entity formed by a plurality 
of individual subjects associated in such a way as to form 
a group ; there remains only the collective subject " (p. 74). 
Society must therefore "be regarded as a personality 
qualitatively different from the individual personalities of 
which it is composed" (p. 54), and altruism becomes 
legitimate in so far as the other person incarnates society. 
For " though society is something different from the 
individual, and though it can be found in its entirety in 
none of us, yet there is not one of us in whom it is not in 
a manner reflected " (Education Morale, p. 93). 

It should be noted, in the first place, that such an ex- 
planation of altruism could equally well be made to 


justify egoism. For, after all, I am as much a reflection of 
society as anyone else, and this inner reflection seems to 
me far more direct than those that are refracted through 
other people's personalities. Apart from this interpreta- 
tion, these passages from Durkheim's writings can have 
only one meaning : what we love in others is not so much 
the individual as the possibility of a bond of affection, not 
so much the friend as friendship. An inhuman formula if 
the moral fact remains external to the individual, and is 
imposed upon him from without, but one of singular 
profundity if the good constitutes the law of perspective 
and the rule of reciprocity which aim at bringing about 
mutual understanding. Then what we seek in the other 
person is the very thing that enables the other person to 
come out of himself while yet remaining most profoundly 

The good, in short, is not, like duty, the result of a 
constraint exercised by society upon the individual. The 
aspiration to the good is of different stuff from, the obedience 
given to an imperative rule. Mutual respect, which consti- 
tutes the good, does not lead to the same conformity 
as unilateral respect, which characterizes duty. In our 
modern societies, no doubt, the difference between the 
two has become dulled, for the content of duty can be 
more and more identified with the good. This is why 
Kantian Ethics, in its form a morality of duty, is in its 
content a morality of autonomous will, and defines the 
good by a universality which rests on reciprocity itself 
(though in Kant's persona,! mentality there are many 
traces of heteronomy and legalism). But in primitive 
communities the opposition is almost complete between the 
whole set of legal prohibitions or taboos laid down by the 
morality of duty and the rules of justice and reciprocity 
which grow up between individuals without always being 
codified. It is only in differentiated societies that, as 
ritual obligations diminish along with conformity, the 
morality of good wins against the morality of duty, and 
becomes transmitted from one generation to another until 


it comes to constitute the actual content of the duties 

In conclusion, the fundamental difficulty of Durk- 
heimism seems to us to be the illegitimate identification 
of constraint and cooperation. In the moral sphere, this 
means that the identification of good and duty is pushed 
too far and, what is worse, that morality is made sub- 
servient to social conformity. There can be no complete 
moral autonomy except by cooperation. From this point 
of view, morality is still a social thing, but society cannot 
be regarded as a completed whole nor as a system of fully 
realized values. The morality of good develops progressively 
and constitutes, in relation to society, a form of ideal equili- 
brium, as it were, which rises above the false and unstable 
existing equilibria which are based on constraint. 


HEIM : II. MORAL EDUCATION. We now reach what 
is to us the fundamental point of the controversy, that is to 
say, Durkheim's ideas on child psychology in relation 
to sociology. For the course of lectures embodied in his 
Education Morale is the most powerful effort made 
on scientific lines to justify the conceptions on which 
traditional education rests. Whereas nearly all child 
psychologists, Stanley Hall, Dewey, Clapar&de, etc. 
speaking, it is true, more from the child's point of view 
than from that of society have denounced the delusions of 
the scholastic method, Durkheim, taking his stand on 
society and approaching the child from that standpoint is, 
on the contrary, extremely conservative. We must there- 
fore test the soundness of his arguments in the light of 
the actual facts of child morality. 

Let us start with a discussion of the principle involved, 
not only because this work of Durkheim's forms a wonder- 
fully coherent whole, but because in many ways he re- 
affirms here what was stated in his earlier writings. 

There are, according to Durkheim, three elements in 
morality : the spirit of discipline, the attachment to 


social groups, and the autonomy of the will. The spirit of 
discipline (V esprit de discipline) is in itself fundamental, 
since morality consists in a body of rules sanctioned by 
society. " To regularize conduct is a fundamental func- 
tion of morality " (p. 30). But there is more in the idea 
of rule than merely the notion of regularity : there is 
authority. " By authority must be understood the ascend- 
ency exercised over us by any moral power that we 
recognize as superior to ourselves" (p. 33). Morality 
is thus a system of commands, and individual conscience 
nothing but the product of the interiorization of these 
collective imperatives. It is also true that discipline, 
far from thwarting the growth and expansion of the 
individual, is the sole condition for the development of 
personality (p. 52). 

The attachment to social groups is no less important. 
Individuals possessing no moral value in themselves, 
the group is the only legitimate end. But here again, 
far from checking individual initiative, this end enriches 
personality : we are all social beings, and there is no real 
antagonism between the individual and society (p. 80). 
Moreover, " What shows that morality is the work of the 
community is that it varies as do communities " (p. 98). 
It is true that beneath all the variations of morality 
there remains a basis that is permanent and unchanging, 
but this is because society, though it evolves, retains 
certain constant features : "A society remains, up to a 
point, identical with itself, throughout the course of its 
existence. Under the changes it undergoes its funda- 
mental constitution is always the same. The ethical 
system which it practises therefore presents the same 
degree of identity and constancy " (p. 121). Finally, 
it should be noted that the spirit of discipline and the 
attachment to a group constitute, in the last analysis, 
one and the same phenomenon. No individual possesses 
authority or prestige in himself. " Society alone stands 
above individuals. From it therefore emanates all autho- 
rity. It is society that bestows upon such and such human 


qualities that sui generis character, that prestige which 
raises the individuals who possess it above themselves " 
(p. 103). 

This unity of discipline and attachment to groups also 
explains the profound identity of duty and good. " Duty is 
morality in so far as it commands ; it is morality conceived 
as an authority which we obey because it is authority, and 
for no other reason. Good is morality conceived as some- 
thing desirable, which entices the will to its service, and 
provokes a spontaneous longing. Now it is easy to see 
that duty is society, in so far as it lays down rales and 
sets limitations to our nature ; whereas the good is also 
society, but in so far as it is a reality richer than our own, 
and one to which we cannot become attached without a 
resulting enrichment of our being" (p. no). These two 
elements of morality are therefore only " two different 
aspects of the same reality " (p. 112), and if it be true that 
our societies lay more stress on good than on duty, and 
that we are losing the sense of collective discipline (p. 116), 
then it becomes the urgent task of education to restore 
the unity of our moral consciousness by reconciling the 
good and duty once again. 

The third element of morality is the autonomy of the 
will. It is contrary to rational morality to impose any- 
thing whatsoever on conscience itself. " It is a rule, not 
only of logic, but of morality, that our reason must only 
accept as true what it has spontaneously recognized as 
such " (p. 123). But the Kantian solution which explains 
autonomy by the rational will reduces obligation to 
" an accidental character, as it were, of the moral law " 
(p. 125). But it is necessary, on the contrary, to justify 
autonomy without disparaging the principle of obligation 
and authority. " Kant has shown better than anyone 
else that there was something religious in the feeling with 
which the moral law filled even the highest reason. Now 
we can have religious feeling only for some being, real or 
ideal, who seems to us superior to the faculty by which 
we conceive it. For the truth is that obligation is an 


essential element of moral precept " (p. 126). Durkheim 
claims to find this reconciliation of autonomy and social 
authority in a comparison with natural science. It is 
only by learning the laws of nature, by using them with- 
out trying to infringe them, that we become free of 
nature. Now " in the moral order there is room for the 
same autonomy, and there is room for no other. Just as 
morality expresses the nature of society, and that society 
cannot be known directly any more than can physical 
nature, so individual reason can no more legislate for the 
world of morals than it does for the material universe. . . . 
But this order which the individual, as an individual, has 
not created, has not deliberately willed, can become his 
by science" (p. 133). In short, autonomy consists in 
understanding the why and wherefore of the laws which 
society lays upon us and which we cannot choose but 

After this analysis, Durkheim asks how we can instil 
the elements of morality in the child. Durkheim's prin- 
ciple is very elastic when he comes to deal with educational 
matters, far more so, indeed, than the sociological doctrine 
would have led one to suppose. On the one hand, the 
child does not of itself possess the elements of morality, 
and we must therefore " inform its nature." But on the 
other hand, " educational action . . . does not work upon 
a blank sheet of paper. The child has a nature of its own, 
and since it is this nature that must be informed, we 
must, if we are to work upon it effectually, begin by try- 
ing to know it " (p. 147). It is these introductory remarks, 
so wise and so much in tune with the contemporary 
educational movement, that have led us to submit Durk- 
heirn's ideas to a searching critical examination. 

Unfortunately, under the influence of a " pre-notion ", 
hard to account for in a sociologist, and especially in one 
so methodical, Durkheim thinks of children as knowing 
no other society than adult society or the societies 
created by adults (schools), so that he entirely ignores 
the existence of spontaneously formed children's societies, 


and of the facts relating to mutual respect. Consequently, 
elastic though Durkheim's pedagogy may be in principle, 
it simply leads, for lack of being sufficiently informed on 
the subject of child sociology, to a defence of the methods 
of authority. 

First question : How to form the spirit of discipline 
in the child ? It should be noted from the first " that the 
mental states which education must produce in the child 
are only virtually present, and exist only in a general 
form that is very far removed from that which they are 
destined to take/' This is particularly true "of the 
spirit of discipline. We may say indeed that not one of 
the elements that compose it is present in its entirety in 
the mind of the child " (p. 148). One of these elements 
is the taste for a regular life. Now the child is all fantasy, 
all restlessness. " The outstanding feature of infantile 
curiosity is that it is unstable and fleeting" (p. 149). 
The spirit of discipline, on the other hand, means modera- 
tion and self-control. Now the child knows no limit to 
its desires, sets no curb to its emotions and instinctive 
tendencies. It is true that in spite of this there are " two 
characteristics constitutive of the infantile mental make- 
up which respond to our influence ; these are i childish 
traditionalism; 2 the child's openness to suggestion, 
especially to imperative suggestion " (p. 153). At this 
point, Durkheim makes the very penetrating observation 
that in spite of his restlessness the child is a " lover of 
routine " (p. 153). The rituals attached to eating, going 
to bed, etc., show the hold which habit has over his 
nature. " Thus the child is at the same time unstable 
and misoneistic " (p. 154). Like primitive man, the child 
is a traditionalist (p. 155). At the same time, everyone 
has noticed its suggestibility as well as the enormous 
ascendency which the example of those around it exercise 
upon its nature. 

It is these features of child psychology which render 
futile all those " oft-repeated arguments as to whether 
the child is born moral or immoral, or whether it possesses 


in itself the positive elements of morality and immorality. 
Put in this way, the problem admits of no definite solu- 
tion. To act morally is to conform to the rules of morality. 
But "the rules of morality are external to the child's 
conscience ; they are worked out apart from him ; he 
only comes in contact with them at a definite point in 
his life " (p. 167). This decisive moment in the making 
of the spirit of discipline arrives when the child goes to 
school. In the family altruistic leanings and feeling 
of solidarity are stronger than duty. At school, on the 
contrary, there must be rules. These rules must be 
cultivated for their own sake. They constitute " an 
instrument of moral education which it would be hard 
to replace " (p. 171). It is therefore the master's business 
to impose them. " Since it is by the master that rules 
are revealed to the child, everything rests with him. 
Rules can have no other authority than that which he 
confers upon them: that is to say, than that of which 
he suggests the idea to the children " (p. 176). The lay 
teacher must be a kind of priest of society. " Just as 
the priest is the interpreter of God, so he is the interpreter 
of the great moral ideas of his time and his country " 
(p. 177). But, on the other hand, the teacher's authority 
must give way before that of the rule, " for the rule ceases 
to be itself if it is not impersonal, and if it is not repre- 
sented as such to people's minds " (p. 178). 

Durkheim is therefore opposed to the form of education 
based on individual interests and free initiative which 
is advanced by the " Activity School " under all its forms. 
He answers Montaigne with the classic argument : " Life 
is not all play ; the child must prepare himself for pain 
and effort, and it would therefore be a disaster if he were 
allowed to think that everything can be done as a game " 
(p. 183). There must therefore be not only a firm discipline 
in schools, but also penalties that will enforce this discipline. 

But what exactly do we mean by a penalty ? Some 
people look upon it as preventive measure. But the 
superficiality of such a justification has been recognized, 


since it attributes to the child or delinquent a capacity 
for prevision which is utterly foreign to their natural 
impulsiveness. Others, again, take a penalty to be an 
expiation. Durkheim, while he rejects the materialistic 
form of the doctrine of expiatory suffering, thinks, never- 
theless, that <c something of this theory should be retained. 
The part of it to be retained is the principle that the 
penalty wipes out the fault, or at any rate makes up for it 
as much as is possible. But the penalty does not possess 
the power of reparation because of the suffering it im- 
plies : for suffering is an evil, and it would be absurd to 
suppose that one evil can compensate for another and 
nullify it" (p. 188). If suffering wipes out the fault, 
this is because " the fundamental function of a penalty " 
is "to reassure those spirits whose faith may indeed 
must have been shaken by the violation of the law, even 
though they were not aware of it, to show them that this 
faith is as justified as it was before, and, especially in the 
case of schools, that it is still held by the person who 
first gave it to the children " (p. 191). Thus a punish- 
ment wipes out a fault in so far as the pain inflicted proves 
to the child that the teacher has taken this misdeed seri- 
ously. So that the essence of a penalty is to symbolize 
blame : " The punishment is only the material sign of 
an inner state ; it is a system of notation, a language, 
by means of which either public opinion or the school- 
master's conscience expresses the feeling aroused in them 
by the censured act " (p. 201). 

As for Herbert Spencer's theory of natural punishment, 
it rests on the illusion that education comes from nature, 
whereas in reality it is a social process. Punishment 
should undoubtedly be regarded as the natural conse- 
quence of the fault committed, but the consequence in 
this case is not the physical result of the guilty actions 
themselves. Punishment is the result of the fault only 
in so far as the latter sets afoot a movement of dis- 
approval symbolized by the former. " The true punish- 
ment, like the true natural consequence, is censure " (p. 205) . 


It should be noted here that coercive in character as 
is this type of pedagogy, Durkheim protests vigorously 
against corporal punishment, and shows himself a very 
subtle psychologist in doing so. Corporal punishment, 
which is unknown to most primitive communities, de- 
veloped with the school and reacted on the family. How 
did it take its rise in the school ? This is " a special case 
of a law which could be formulated as follows. Whenever 
two peoples, two groups of individuals belonging to 
different levels of culture, are brought into continuous 
contact with each other, certain feelings will develop 
which will make the group which has or believes itself to 
have the higher culture tend to do violence to the other 
group 1 ' (p. 219). Thus, faced with his pupils, the master 
can easily be led to violence, or even megalomania : 
since the school " has by nature a monarchical form, it 
easily degenerates into despotism " (p. 224). 

These considerations on discipline apply equally to 
the second element of morality : how to form in the child 
a normal attachment to the social groups that surround 
him. The psychological origin of this part of our moral 
life is to be sought for in our faculty for sympathy and in 
our altruistic and disinterested tendencies. Now the 
child is " neither purely altruistic nor purely egoistic, but 
both at the same time, though to a far lesser extent than 
we are " (p. 260). It will therefore be enough to extend his 
social circle progressively in order to educate him. Before 
going to school he has known family life and the society of 
friends, in which altruism is strengthened by ties of blood 
and comradeship. And in order to prepare him for political 
society which does not possess these characteristics, 
school is the obvious intermediary. Civic and historical 
teaching are sufficient to initiate the child to the values 
attaching to adult society. 

Such, then, in broad outline is Durkheim's moral 
pedagogy. We cannot enter upon a discussion of a work 
so sincere, so lofty in inspiration as that which we have 
just summarized without a feeling of profound respect for 


the memory of its author. But such is the seriousness 
of the questions at stake, that we must not hesitate to 
examine these theories of Durkheim's in detail and in a 
spirit of complete freedom. The greatest tribute we can 
pay to his vigorous scientific spirit is to forget for a 
moment his immense authority. 

Now, the doctrinal foundation for Durkheim's pedagogy 
is so contrary not only to the apparent achievements of 
present-day child psychology, but also to all that seems 
to have been established by the new pedagogy (and this 
is more serious, for the experiments of pedagogy will 
always be more conclusive than those of the laboratory), 
that we find it impossible to accept Durkheim's con- 
clusions as they stand. In spite of the force and fertility 
of its methods and principles, Durkheim's sociology 
carries with it a fundamental difficulty that, namely, 
which we pointed out a little while ago, and of which 
the effects are now apparent in child psychology. Not con- 
tent with showing that the main key to human psychology 
is to be found in social life, Durkheim tries to make of 
society a whole, a " being ", and this realism, like all 
realism, has given rise to those antinomies which a metho- 
dical relativism alone can avoid. There are no more 
such things as societies qua beings than there are isolated 
individuals. There are only relations ; these relations 
must be studied simultaneously from outside and from 
inside (there being no possible conflict between psychol- 
ogy and sociology), and the combinations formed by them, 
always incomplete, cannot be taken as permanent sub- 
stances. It is therefore impossible to subsume under one 
concept the many and diverse effects which social life 
exercises upon the development of the individual. The 
unwarranted identification of constraint and cooperation 
is therefore what we shall once again have to discuss, for 
this identification vitiates the whole of Durkheim's 
pedagogy, as it vitiates his ethics. Let us then examine 
from this point of view the three elements which Durk- 
heim recognizes in morality. 


The spirit of discipline, it must be agreed, constitutes the 
starting-point of all moral life. There must be, not only 
a certain regularity of behaviour, but rules, and rules 
clothed with sufficient authority. No one will seriously 
deny that this is the price to be paid for the development 
of personality. We can also agree with Durkheim when 
he sees in the individual as such nothing but a creature of 
habit and routine. Thus it is social life that elaborates 
rules properly so called, and this point of view receives 
full confirmation from our enquiries : all the rules followed 
by children in all the domains we examined are due to 
social relations. 

Only and herein resides the whole problem of educa- 
tion as of morality, and even of logic is there only one 
type of authority and one type of rules ? Durkheim 
settles the question almost without discussing it, and 
without seeming to suspect that alongside of the social 
relations between children and adults there exist social 
relations that apply distinctly to the groups which children 
form among themselves. The passages in which Durk- 
heim commits this most disastrous petitio principii are pro- 
bably among the most dogmatic in the whole of his works. 
All authority comes from Society with a big S (" la " 
societ^, p. 103) ; the schoolmaster is the priest who acts 
as an intermediary between society and the child (p. 177) ; 
everything therefore rests with the master (p. 176), and 
rules are a sort of revelation (p. 176) which the adult 
dispenses to the child. 

Now, if psychological observation gives one indubitable 
result, it is that, far from limiting himself to the rules 
laid down by his parents and teachers (those which he 
very often observes least closely), the child ties himself 
down to all sorts of rules in every sphere of his activity, 
and especially in that of play. These rules are no less 
social, but they rest on different types of authority. 
Now the question that educationalists asked themselves 
was whether these rules distinctive of children's societies 
could not be utilized in class, and their experiments 


few in number and little known though they were towards 
1902-3 and 1906-7 (the dates of Durkheim's lectures), 
have given birth to a system of moral pedagogy, called 
sdf -government, which is at the opposite pole from the 
Durkheimian pedagogy. 1 A fundamental problem of 
psychology thus comes to be formulated as follows : 
Do these different kinds of rules obey the same psycho- 
sociological laws, or do they not ? This is the question 
which has occupied us throughout this book, and our 
answer has been in the negative. 

For it seemed to us that there existed at least two 
extreme types of rules and of authority rales due to 
unilateral respect, and rules due to mutual respect. 
Durkheim tells us that the individual as such has no autho- 
rity. But even if this statement is well founded (which 
we shall consider in connection with M. Bovet's theory), 
the fact remains that the playmate, or older child, con- 
stitutes, like the schoolmaster or the parent, a reflection, 
not of Society (with # big S), but of this or that society 
thought of by the child, and that the respect felt for 
a fair-minded companion in a match of marbles is 
different from that called forth by an adult. And these 
two types of rules lead to opposite results. The rule of 
constraint, which is bound up with unilateral respect, 
is regarded as sacred and produces in the child's mind 
feelings that are analogous to those which characterize 
the compulsory conformity of primitive communities. 
But this rule of constraint remains external to the child's 
spirit and does not lead to as effective an obedience 
as the adult would wish. Rules due to mutual agreement 
and cooperation, on the contrary, take root inside the 
child's mind and result in an effective observance in the 
measure in which they are incorporated in an autonomous 

If this distinction is based on fact we may in the very 
name of sociology ask the following question. Contemporary 

1 See, in particular, Ad. Ferriere's fine book, UAutonomie des Ecoliers, 
Collection des Actualit6s PSdagogiques, NeucMtel et Paris. 


civilized society, that, namely, 'to which we are seeking 
to adapt the child, is tending more and more to sub- 
stitute the rule of cooperation for the rule of constraint. 
The essence of democracy resides in its attitude towards 
law as a product of the collective will, and not as some- 
thing emanating from a transcendent will or from the 
authority established by divine right. It is therefore the 
essence of democracy to replace the unilateral respect 
of authority by the mutual respect of autonomous wills. 
So that the problem is to know what will best prepare 
the child for its future task of citizenship. Is it the habit 
of external discipline gained under the influence of uni- 
lateral respect and of adult constraint, or is it the habit 
of internal discipline, of mutual respect and of " self- 
government " ? It may be, of course, that only those 
who have gone through the external discipline imposed 
by a master will be capable later on of any inner discipline. 
This is the commonly accepted view, but it requires to 
be verified. The proof, however, would not be an easy 
one to establish, for considering the large number of 
people who reject all discipline as soon as they have 
escaped from school and home ties, or who for the rest of 
their lives are capable only of external discipline and legal 
morality, it may very well be that it is in spite of adult 
authority, or in spite of certain kinds of adult authority, 
that the best of our young people sooner or later adopt 
a disciplined way of living. 

For ourselves, we regard as -of the utmost importance 
the experiments that have been made to introduce demo- 
cratic methods into schools. As Foerster x has so ably said, 
it is unbelievable that at a time when democratic ideas 
enter into every sphere of life, they should have been 
so little utilized as instruments of education. If one thinks 
of the systematic resistance offered by pupils to the 
authoritarian method, and the admirable ingenuity 
employed by children the world over to evade disciplin- 
arian constraint, one cannot help regarding as defective 

1 F. W. Foerster, UEcole et le Caractire. Fr. translation by Bo vet. 


a system which allows so much effort to be wasted Instead 
of using it in cooperation. 

One should study, for example, the evolution of a 
born educator like Sanderson to see how this headmaster, 
who was at first a partisan of strict authority and even 
of corporal punishment, ended by introducing democracy 
and the system of collaboration into his boarding-school. 1 
We therefore do not at all agree with Durkheim in think- 
ing that it is the master's business to impose or even to 
"reveal" rules to the child. A "priest" is the last 
thing a schoolmaster should be : he should be an elder 
"collaborator, and, if he has it in him, a simple comrade 
to the children. Then only will true discipline come into 
being discipline that the children themselves have willed 
and consented to. Every educationalist who has really 
made the experiment has found that this is what actually 
happens. The sense of a common law which, as we have 
shown in connection with the rules of a game, is possessed 
by children of 9-12, shows clearly enough how capable 
is the child of discipline and democratic life, when he is 
not, as at school, condemned to wage war against authority. 

It is true that the problem of discipline is bound up with 
that of functional education as a whole. Autonomous and 
Inner discipline can exist in a class only to the extent that 
the work enlists the major part of the child's spontaneous 
initiative and activity. Interest being, according to 
Dewey, the participation of the " Ego " in the work done, 
it is obvious that it is necessary to the elaboration of a 
discipline proper to a system of autonomy. 

Only the Activity School, i.e. that in which the child is 
not made to work by means of external constraint, but 
where he works of his own free will (from the psychol- 
ogical point of view the work done in these two cases is 
completely different) only the Activity school is able to 
realize cooperation and democracy in the class-room. 
No.w it is precisely the principle of the Activity School 
that Durkheim is disputing in his refutation of Montaigne 

i H. G. Wells, A Great Headmaster. 


and Tolstoi, and in doing so lie makes use of the traditional 
objection that life is not a game, and that it is not by 
playing that the child will learn to make efforts. 

Nowadays, after the many Activity Schools that have 
been established in Europe and America, it would no 
longer be possible to use such an argument. Every 
practising teacher who has made the experiment in 
favourable conditions, has observed and anyone who 
watches a child outside school hours will have noted the 
same thing that if a child is interested in what he does 
he is capable of making efforts to the limits of his physical 
endurance. The problem can therefore be stated in the 
same terms as above, but under the following general 
formula : Who is the man who will be capable of the 
greatest energy in precisely those circumstances where 
life is not a game ? Is it the man who has best learnt 
as a child to make this voluntary and spontaneous effort, 
or the man who has never worked except in obedience 
to a command ? The present writer has in this connection 
some very definite recollections. In his class a little 
class in a little town in Switzerland there were, as there 
are in all classes, a few thoroughly lazy boys, a few sound 
and conscientious boys, and a few who were only moderately 
good pupils in school, but who went in for " interesting 
things " at home chemistry, the history of aviation, 
zoology, Hebrew, anything you like except what was 
on the curriculum. The conscientious boys, who did not 
take school life as a game have, some of them, become 
civil servants, while others have filled minor academic 
posts, and so on ; they cannot really be regarded to-day 
as models of active energy. The same thing happened 
to the lazy ones, when they did not disappear altogether. 
As to the moderately good, after being told throughout 
their school life that " If only you spent on your home- 
work a quarter of the time you spend on your personal 
pursuits, you would do exceedingly well", they ended by 
giving up all their time to these personal pursuits, and 
are now very sorry they were not able to extend this 


method to many branches of learning in which they have 
remained ignorant. It should be added, however, that 
among our masters were some who not only understood 
but encouraged and utilized this turn of mind and who, 
like elder comrades, as it were, really enriched our fund 
of knowledge because they had the tact to guess every- 
thing and impose nothing. 

In short, effort, like every other form of behaviour, 
requires to ripen, and its primitive forms, though very 
different from the final product which alone receives the 
sanction of adult morality, 1 may well be indispensable 
to the normal development of the individual. It is there- 
fore not wasting a child's time to let him acquire by 
himself the habit of work and of inner discipline. In 
the moral as in the intellectual domain we really possess 
only what we have conquered by ourselves. And if a 
child is to achieve a desire for work and the habit of 
making efforts, we must take his interests and the special 
laws of his activity into account, and we must not impose 
upon him from the first a system of methods and ways 
too much like our own. Besides, do we ever in our own 
lives carry out an effort that is comparable to that which 
we require of children ? The tedious obligations laid 
upon us by ordinary life, or the painful tasks called forth 
by exceptional circumstances never set our effort going 
unless we accept them, and we can only accept them by 
really understanding them. And this is something very 
different from that often meaningless obedience which 
we claim to be a right preparation for life and which 
as often lays the foundations of revolt or passivity. 
This does not mean that the best method of education is 
to let children do exactly as they like, or that the in- 
dividual's own instincts will lead him to effort, work, 
and discipline. For work and discipline to come into 
being there must be and on this point Durkheim is 
perfectly right an organized social life. But the founda- 

1 ClaparMe (Educational Experiment and Child Psychology) has 
written conclusively on this point. 
2 A 


tions of this social life can be laid without despotism or 
constraint. The school, according to Durkheim, is a 
monarchy founded on divine right. We have seen that 
children are capable of democracy. These youthful 
tendencies should be utilized and not lost or set up against 
adult authority, as is so often the case in school life or 
in the dreams of adolescence. 

Thus the whole question of punishment requires re- 
vision. According to Durkheim expiatory punishment is 
necessary, because it is a symbol of censure. But we have 
here at least two points that require closer investigation : 
who should impute the blame, the master or the com- 
munity, and is it really necessary that blame should be 
symbolized by expiatory suffering ? 

As to the first point, educationalists who have applied 
self-government have been gradually led to give up 
adult punishment and to allow free play to retributive 
justice as exercised by the tribunals constituted by the 
children themselves. One need only read the reports 
of these children's debates and judgments to form an 
opinion on the immense progress represented by this 

As to the psychological character of punishments, must 
we admit without argument that expiatory suffering is 
necessary in order to symbolize blame and reassure 
people's consciences ? Even if the symbol is necessary 
for those who are incapable of understanding the reasons 
for which blame is imputed, it loses all significance for 
those who understand these reasons. Moreover, if the 
punishment has no other relation with the act it 
punishes than this relation of being a simple sign, it will 
always be " arbitrary " in its content. Now this is just 
what the rational mind cannot allow, and what those 
people try to avoid, who want to make the punishment the 
natural result of the action. Durkheim, it is true, objects 
to such a view on the ground that the natural consequence 
of the action is the disapproval and consequently the 
punishment which it provokes. But this is to give a very 


wide meaning to the term " natural ". One might as 
well say that the natural consequences of disobedience 
are the master's anger and the resulting blows he delivers, 
which is the last thing Durkheim would approve of. By 
" natural " Durkheim means a well-defined social reaction, 
and one that has been defined, by means of judgments 
of moral value. In that case would it not be more natural 
to be content to impute blame without symbolizing it 
by arbitrary suffering, or to content oneself with punish- 
ments having the approval of the morality of mutual 
respect, such as punishment by reciprocity ? In short, 
on this point as on so many others, Durkheim strikes us 
as trying artificially to maintain the unity of the morality 
of authority with rational morality. Hence his efforts to 
safeguard the idea of expiation which the morality of co- 
operation can only denounce as materialistic and immoral. 

If we consult the children themselves and this is the 
only possible course for a psychologist who wishes to 
observe the laws governing the evolution of moral con- 
sciousness and not merely to lay down an a priori ethic 
we find that the punishment that seems most fair at the 
stage of cooperation is that which is based on the idea 
of reciprocity. Now, to return to Durkheim's argumenta- 
tion, is not this type of punishment sufficient to " reassure 
people's consciences", and to "symbolize" blame? 
Punishment by reciprocity, instead of being an arbitrary 
symbol, is " motivated " by its own content. As for the 
childish conscience, it will be all the more "reassured" 
with regard to the authority of rules, owing to the fact 
that in the case of punishment by reciprocity this 
authority emanates, not from adult revelation but simply 
and solely from mutual and autonomous respect. 

We come now to the attachment to social groups. 
Durkheim very reasonably remarks in this connection 
that the child is neither fundamentally selfish, as has so 
often been held, nor purely altruistic, as has also been 
maintained, but both egoistic and altruistic, though to 
a lesser degree than the adult. This state of things is, 


in our opinion, bound up with the phenomenon of childish 
egocentrism in which we have tried to find the key to 
the psychological phenomena characteristic of the child 
mind. For in childhood, society and the individual are 
as yet undissociated, so that while the child is in the 
highest degree suggestible to his adult surroundings, whether 
through imitation of example or through constraint, he 
yet contrives in a purely unconscious manner to reduce 
everything to his own point of view. This state of things 
combines very easily with the psychological attitude 
that characterizes conformist societies, and we saw, in 
connection with the game of marbles, how easy is the 
synthesis of egocentricity and constraint ; but it is 
strongly opposed to cooperation, in so far as the latter 
implies personalities that are both conscious of themselves 
and able to submit their point of view to the laws of 
reciprocity and universality. The whole problem is 
therefore how to take the child out of his egocentricity 
and lead him to cooperation. 

Durkheim remarks, in this connection, that the only 
societies known to the child are the family and his friends, 
whereas the aim of social education is to prepare him for 
his country and for the more specifically human feelings. 
The school is therefore the necessary link. All this is 
perfectly true, but here again we are faced more irrevo- 
cably than ever with the problem of constraint and co- 
operation. The modern ideal is cooperation dignity of 
the individual and respect for general opinion as elabo- 
rated in free discussion. How are we to bring children 
to this spirit of citizenship and humanity which is postu- 
lated by democratic societies ? By the actual practice of 
cooperation, as soon as this is psychologically possible, 
by what Foerster has so happily called " democracy at 
school ", or by a verbal initiation into the mechanism of 
adult society ? Will the lessons in history and sociology, 
will training in citizenship and instruction in the elements 
of law, have the least effect on the youthful mind if the 
school still remains the system of monarchical authority 


favoured by Durkheim ? For who will turn out to be the 
best citizen, the " sport ", the " chic type " of youthful 
communities, or the pupil who can give the best answers 
in his history and civic instruction examinations ? The 
question is worth considering. 

One last point remains ; its educational application 
is not dealt with in the MSS. which Durkheim has left 
of his course of lectures, but their author has studied it 
in connection with the elements of morality, we mean 
the autonomy of the will. The position which Durkheim 
adopts with regard to Kantian Ethics is extremely sug- 
gestive, and his penetrating remarks on the subject reveal 
more than all the rest of his work, the strength and the 
weakness of his own doctrine. According to Kant, the 
autonomy of the moral will is due to its rational character, 
the feeling of obligation belonging to duty as such being 
the result of sensibility. Durkheim points out, quite 
rightly in our opinion, that in this case the obligation 
belonging to pure duty becomes " in a way an accidental 
feature of the moral law "' (p. 125). Now, according to 
Durkheim, everything proves " that the moral law is 
invested with an authority that claims respect even from 
reason " (p. 125), and that it is therefore the work of an 
entity, real or ideal, which is superior to the faculty that 
has conceived it" (p. 126). And as this higher being is 
no other than society itself, the only possible autonomy 
is free submission on the part of individual reason to the 
laws of society. 

The enormity of such a solution is obvious. Because 
Durkheim seeks to assimilate to one another constraint 
and cooperation, duty and good, he actually comes to 
identify the two most antithetical conceptions of obliga- 
tion the heteronomous submission of reason to the 
" higher entity " and the necessity residing within reason 
herself. That obligation in the first sense of the term, i.e. 
duty in so far as it implies unilateral respect and the 
feeling of authority as such, that obligation in this sense 
should be external to morality, is the logical outcome of 


the Kantian critique, but it is also the natural conse- 
quence of the morality of cooperation. This will shock 
only those who are incapable of experiencing for them- 
selves the higher and purely immanent feeling of obligation, 
which is the product of rational necessity. " Duty/' says 
Durkheim, " is morality in so far as morality commands ; 
it is morality conceived as an authority which we must 
obey, because it is authority and for this reason alone " 
(p. no. The italics are ours). Well, if this is duty, all 
one can say is that it is quite incompatible with the 
morality of cooperation. Besides, even if Durkheim the 
sociologist would have found himself embarrassed by such 
a conclusion, Durkheim the man, Durkheim the free and 
generous spirit, showed by the whole of his career that 
the latter was his real faith. 

Cooperation, mutual respect, therefore imply something 
more than the illusory autonomy described by Durk- 
heim ; they postulate the complete autonomy of reason. 1 
When Durkheim reminds us that the individual is unable 
of himself to create morality, this by no means implies 
that the person (i.e. the individual in so far as he submits to 
the norms of reciprocity) is not free to judge of everything 
by his reason alone. Nothing could be truer than to say 
that autonomy presupposes a scientific knowledge of social 
as of natural laws and the ability to recognize these laws at 
work. But social laws are unfinished and their progres- 
sive formation presupposes the unfettered cooperation of 
personal reason. The autonomy of reason has therefore 
nothing to do with individual fancy, but it stands in direct 
contradiction to the idea of external authority recognized 
as such. As Rauh has so penetratingly shown, to be 
personal is to " situate oneself ", which does not mean 
that one must not first " inform oneself ". In order to 
foster autonomy in the child it is therefore very useful 
to " inform " him scientifically, but to do this it is not 
sufficient to make him submit to the dictates of adult 

i On this point, see amongst other works M. D. Parodi's excellent 
book, Le PyobUme moral et la Pens be contemporaire, 3rd e<L, 1930. 


society and to explain from outside the reasons for tMs 
submission. Autonomy is a power that can only be con- 
quered from within and that can find scope only inside 
a scheme of cooperation. 

In conclusion, then, we find that on every point of 
moral sociology and pedagogy Durkheim's teaching 
presents both an optimum et pessimum. Profoundly 
true in its conception of moral facts as social and bound 
up with the structural and functional development of 
collective groups, it overlooks the essential difference 
existing between cooperation and constraint. Hence 
Durkheim's illusion that an education which makes use 
only of unilateral respect can lead to the results that are 
peculiar to the morality of mutual respect. Hence, in 
moral psychology, his confusion between the heterono- 
mous character peculiar to pure duty and the quality 
of fundamental autonomy which belongs to the good as 
such. Hence, finally, in his general sociology the un- 
warranted identification of the equilibrium of fact con- 
stituted by constraint and that other, ideal equilibrium 
still social though in another sense of the term which 
is constituted by cooperation, the limit and norm of every 
human group that has ever come into being. 

4. M. PIERRE BOVET'S THEORY. The premises on 
which M. Bovet bases his theory are at once very near 
to and very far removed from those of Durkheim. Like 
Durkheim, M. Bovet refuses to account for the moral 
feelings by means of the psychological phenomena be- 
longing to the individual as such. At any rate, in speaking 
only of individuals, M. Bovet is of opinion that an isolated 
being would never develop in himself a sense of moral 
obligation, a feeling which does not admit of being reduced 
to those facts of adaptation, of habit, or of instinctive 
afiectivity to which it has been compared. Like Durk- 
heim again, M. Bovet rejects the Kantian attempt to 
interpret respect as an effect of rational law, and en- 
deavours, on the contrary, to explain rational rules by 


respect, and respect by the empirical conditions of 
social relations themselves. Finally, like Durkheim, 
M. Bovet carefully distinguishes between the feeling of 
the good and the sense of duty. Only, where Durkheim 
speaks of society as of a thing which exercises pressure 
upon the individuals, M. Bovet is thinking only of the 
relations existing between individuals. Moreover, it is 
sufficient in his view that there should be contact between 
two individuals for one to respect the other, and for those 
moral values to appear which are born of this respect. 
The analysis of the facts of conscience seems to him, there- 
fore, to fulfil all the demands of moral psychology, without 
involving the necessity for an initial sociological enquiry. 
It is this difference of method rather than any diver- 
gence of doctrine which seems to us best to define the 
contrast between Durkheim's theory and that of M. 

But if, in view of recent work done on the subject, 
a sort of parallelism be allowed between the sociological 
method, which describes facts from outside, and the psycho- 
logical method, which still regards social facts as funda- 
mental, even if they are interpreted in terms of conscious- 
ness, the opposition between Durkheim and M. Bovet 
reduces itself to very little. It is with those theories that 
try to explain the moral law as something inborn, or as 
due to the individual's own experience alone, that the 
sociological point of view is incompatible. Once you 
admit, with M. Bovet, the necessity for contact between 
at least two individuals in order to constitute an obli- 
gatory rule, questions of detail lose their importance. 
Psychologists and sociologists can labour together in the 
erection of a science of the facts of morality. 

It is in this spirit of cooperation that we should like 
to discuss the ideas of M. Bovet after those of Durkheim. 
If the former seem to require the acceptance of the latter 
or of as much as we have retained of the latter the 
converse also seems to be true, and M. Bovet's method 
seems to us to be indispensable to anyone who wishes to 


formulate a problem of moral psychology in experimental 
terms. The success that has attended sociological re- 
search has no doubt very nearly stifled the interest that 
was taken about twenty years ago in attempts made at 
establishing a systematic analysis of the facts of will 
and consciousness of rules, and this probably explains 
why the new departure made by M. Bovet did not create 
more of a sensation. But if one compares the conclusions 
of a remarkable and too little quoted article of his 1 with 
the poor and unreliable results of the deliberate intro- 
spection then current in the study of will, one cannot 
help thinking that here was a point of view which, except 
that it was not worked out on a large scale, deserved to 
be placed on the same level as Durkheim's. For our 
part, if the observations contained in the present work 
survive independently of the provisional interpretations 
which have served as a framework for them, we do not 
hesitate to declare that this article of M. Bovet's was the 
true begetter of our results. 

What are the conditions of the feeling of obligation 
in the mind ? Limiting the question to the psycho- 
logy of duty which he opposes to that of the feeling of 
good, M. Bovet answers that two conditions are necessary 
and their union sufficient for the fact of obligation to 
take place : that on the one hand the individual should 
receive commands, and on the other, that he should 
respect the person who gives him these commands. No 
commands, no rules, and consequently no duties ; but 
without respect, the rules would not be accepted, and the 
rules would have no power to compel the mind. 

,We can deal briefly with the first point, since it 
is, common to Durkheim and Bovet. Its significance 
comes to this, that in a given individual there exist 
no duties that emanate from himself. For in the 
first place, moral rules or duties cannot be reduced to 
factors of purely individual psychology. Habit, for 

1 P. Bovet, "Les conditions de 1'obligation de la conscience/' Annie 
psychoL, 1912. 


example, does not suffice to explain duty, since it is not 
accompanied by a feeling of obligation ; there are even 
habits against which we feel it is our duty to struggle. 
But in that case could not spontaneous duties and even 
feelings of remorse arise that were free from all external 
influence? The psychology of sexual phenomena has 
given occasion for such hypotheses. But when one ex- 
amines matters more closely one realizes that without 
the commands coming from those around him and in 
this domain no one can escape from innumerable explicit 
and implicit commands the individual's mind would 
never add any moral judgment to the purely physiological 
impressions of pleasure or of depression. In the second 
place, and resulting from this, we cannot regard as primitive 
any lines of conduct which the individual imposes on him- 
self, and which in reality take their rise in inter-individual 
behaviour. Thus a command or a suggestion which one 
gives oneself inwardly may be the starting-point of a 
feeling of obligation. But these are either mere replicas 
or based by more or less remote analogy on directions and 
suggestions connected with commands that have been 
received. For instance, I decide to start work at such 
and such a time ; but where would the feeling of obliga- 
tion, or, in case of default, the feeling of dissatisfaction 
which accompany my actions come from if I had not 
learnt in my childhood to work at fixed hours and to 
keep rny engagements ? 

As to the second point, this is where M. Bovet's 
originality comes in. Commands given by those in the 
child's immediate surroundings do not seem to him to 
bring about of themselves any sense of obligation in the 
child mind. The child as well as the adult sometimes 
hears rules formulated at which he is content to smile, 
or which even cause his moral judgment to revolt. How 
is such a phenomenon, in which sociologists will claim to 
see the clash of heterogeneous constraints, psychologic- 
ally possible? Because, says M. Bovet, the command 
does not compel of itself. If it is to arouse in the mind the 


feeling of duty, the command must come from an 
individual who is respected by the subject. It matters 
not whether this individual has himself created the 
contents of the command or whether he simply conveys 
a command as he has received it. The main thing is 
that he should be respected by the subject. Then, and 
then only, will the command produce a sense of duty in 
the latter's mind. 

Thus the conception which M. Bovet has formed of 
respect is equally removed from that urged by Kant 
and that expounded by Durkheim, In Kant's view 
respect is not an ordinary feeling arising like any other 
under the influence of a person or a thing. Its appearance 
is occasioned by the moral law in a manner which is 
inexplicable, and if we feel respect for certain individuals, 
we do so, in so far as they incarnate, as it were, this very 
law. According to M. Bovet, on the contrary, the law 
is not the source of respect. It is respect for persons 
which causes the commands coming from these persons 
to acquire the force of law in the spirit that feels respect. 
Thus respect is the source of law. In Durkheim's view 
as in Kant's there is no respect for individuals ; it is 
in so far as the individual obeys the rule that he is 
respected. But this rale, far from simply emanating from 
reason, results as does respect itself, from the authority 
of the group. In a sense, then, law is the daughter of 
respect, but of the respect of the individual for the group. 
To this M. Bovet answers that if in adult society respect 
for the man and respect for the rule are in fact indis- 
soluble, in the child the former can be seen to precede the 
latter. It is a fact that the child in the presence of his 
parents has the spontaneous feeling of something greater 
than and superior to himself. Thus respect has its roots 
deep down in certain inborn feelings and is due to a sui 
generis mixture of fear and affection which develops as a 
function of the child's relation to his adult environment. 

It should be noted, in this connection, that this attitude 
of the child towards his parents does not only in M. 


Bovet's opinion, explain the genesis of the sense of duty. 
In filial piety we have the psychological source of the 
religious sense *. For in virtue of Ms very respect, the 
young child attributes to his parents the moral and 
intellectual qualities which define his idea of perfection. 
The adult is omniscient, omnipresent, just and good, 
the source both of the uniformities of nature and of the 
laws of morality. Naturally, the child does not give spon- 
taneous expression to such a belief, for it is unnecessary 
for him to formulate and impossible for him to codify the 
" pre-notions " which are a matter of course to him and 
which condition in all their detail his moral judgment and 
his conception of the world. But, as M. Bovet has rightly 
remarked, the intensity of certain crises in a child's life 
is sufficient to show how firmly rooted were the implicit 
attitudes which circumstances have thus undermined. 
The discovery of a fault in the behaviour of his parents 
will completely upset the child's confidence. The dis- 
covery of an intellectual failing or of some unforeseen 
limitation of the powers of the adult will jeopardize his 
faith in a world order. It is then that the primitive 
filial sentiments, and in particular the demand for in- 
tellectual and moral perfection, may be transferred to the 
ideal beings which the collective conceptions of the day 
suggest to the religious consciousness of the individual. 

But this is not the whole matter. If at first the adult 
is a god for the child, and if the commands coming 
from the parents suffice to establish that consciousness of 
duty which most religions have identified with the divine 
will, the fact remains that reason plays a part in the 
constitution of the moral ideal. For how are we to explain 
the genesis of personal conscience if originally everything 
is heteronomous ? M. Bovet suggests the following solu- 
tion. On the one hand, reason works over moral rules, 
as she works over everything, generalizing them, making 
them coherent with each other, and above all extending 
them progressively to all individuals until universality 

* P. Bovet, op. tit.. The Child's Religion. 


is reached. Thus, whoever receives a command draws from 
it logical consequences which apply even to the person who 
issues the command. On the other hand, there is bound 
to be in the course of mental development a certain clash 
between the various influences received. Commands 
cut across and more or less contradict each other, and 
the more numerous the individuals respected the more 
divergent obligations will the respecter have to reconcile 
with each other. In this way reason cannot choose but 
introduce the necessary unity into the moral conscious- 
ness. It is through this work of unification that the sense 
of personal autonomy comes to be conquered. 

It is by referring back to these same processes that 
M, Bovet disposes of the objection which has been made 
to Ms as to Durkheiin's theory. If, it has been said, 
everything comes to the child from outside, and if moral 
rules are at the start nothing but commands coming from 
the surrounding authority, how does the " good " come 
to free itself from mere custom, and how does " good " 
respect come to be contrasted with " bad " respect ? So 
long as the influences which the child is under are limited, 
admits M, Bovet, it goes without saying that the subject's 
ideal cannot go beyond that of those who inspire it. But 
as soon as the influences begin to cut across each other, 
reason establishes hierarchies and progress becomes pos- 
sible in relation to the commands originally received. 

Finally, it should be recalled that these developments 
apply only to the consciousness of duty. Side by side 
with the morality of duty M. Bovet upholds the claims 
of the feeling of the good, though he does not attempt to 
explain it. The existence of this inner ideal peculiar to 
the idea of good is what in the last analysis guarantees 
the endurance of the autonomy of conscience : " We shall 
compel ourselves to see clearly the grounds of our respect ; 
criticizing it in order to establish in ourselves orders of 
value, and ranking our respects in accordance with these. 
When the question of respect presents itself to us, we will 
survey critically our sentiments, instinctive or habitual, in 


the name of our ideal ; asking ourselves what it is that we 
respect in the object of our respect, and repudiating the 
lower forms of respect/' (The Child's Religion, p. 155). 

Such, in broad outline, is M. Bovet's theory. Before 
placing it alongside of the results of our enquiries, let 
us try to state more precisely wherein it completes and 
wherein it corrects the doctrine advanced by Durkheim. 

It will be well, in this connection, to introduce into 
sociology the luminous distinction made by F. de Saus- 
sure in linguistics between the static or synchronistic 
point of view and the genetic or diachronistic point of 
view. For from the static point of view the Durkheimian 
theory of duty is unassailable, and to describe the facts 
in terms of inter-individual psychology would only com- 
plete the sociological description without altering it. 
More exactly, these two descriptions are parallel, or 
imply one another. As against this, however, Durkheim 
has drawn from his profound static analyses a genetic 
sociology that was bound to be hypothetical, and it is 
on this point that psychologists and sociologists risk 
having to part company. 

Moral authority and respect, claims Durkheim, can 
exist only in so far as the whole of a group exercises 
pressure on the individual minds. Society, as such, is 
therefore the origin of respect and obligation, and the 
relation between two individuals is insufficient to explain 
the genesis of any of the facts of duty. When, therefore, 
one individual respects another he does so only in so far 
as the other is invested with social authority. Parents 
and teachers have power over the child because they 
embody the morality of the group. Against this rather 
hazardous genetic theory M. Bovet sets up the facts of 
child psychology : it is man as such that the child respects 
in his parents, and if the child accepts the morality of 
the group, he does so in so far as it becomes embodied 
in respected beings. 

From the static point of view, the point of view of the 
pennanence and functioning of social phenomena, this 


problem, which recalls that of the river and its banks, 
is of no importance whatsoever. The great majority of 
the commands prescribed by the adult for the child are 
the actual rules of the morality of the group, and as these 
rules have fashioned the personalities of the parents 
before being handed on to the children it is idle to ask 
whether the child respects the persons in so far as they 
are subject to the rules, or the rules in so far as they are 
incarnated in these persons. 

But from the genetic point of view it is very im- 
portant to determine whether one can speak of a social 
relation where only two are involved, or whether an 
organized group is necessary to the constitution of the 
elementary moral feelings. Now on this point M. Bovet's 
argument seems to us conclusive. In the first place, there 
seems to be no doubt that the feelings of authority and 
respect appear during the first two years, before language, 
as soon as the little creature has discovered in the big 
person someone like himself and yet infinitely greater than 
himself. The feelings compounded of sympathy and fear 
resulting from this situation explain why the child accepts 
the examples and, as soon as he has mastered language, 
the commands coming from his parents, and why, to the 
simple fact -of imitation, there comes so early to be added 
the feeling of rules and obligation. Since all the writers 
on the subject have thought of respect as connected with 
rules, the early impressions we have just mentioned cannot 
be classed anywhere except among the facts of respect. 
Now, as far as the earliest of all rules are concerned, those, 
namely, that relate to eating, to sleeping, to cleanliness, 
and other activities of the infant, there is no room for any 
doubt. It is not because they are current in the social 
group and in all civilized families that the baby accepts 
them, it is because they are imposed upon him by grown- 
ups who are both attractive and formidable. In the second 
place, during the years that follow, the most superficial 
observation will show that the child obeys particular 
as well as general commands. If Durkheim were right, 


only the rules observed by the parents themselves and 
thus emanating from the group before being transmitted 
to the children would be respected by the latter. Now, 
this is simply not the case. When I ask my children not 
to touch my work-table, this rule becomes one of the laws 
of their universe without my being subject to it myself 
or appearing before them as the priest of society. They 
respect this command simply because I am their daddy. 

We have laid some stress upon these genetic con- 
siderations not, we repeat, because they have great im- 
portance for the mechanism of duty in a given society 
that is already completely organized, but because they 
justify the method we have adopted throughout this 
book. Society begins with two individuals, as soon as 
the relation between these two individuals modifies the 
nature of their behaviour, and all the resulting phenomena 
can be described just as well in terms of inter-individual 
consciousness as in the general terms of sociology proper. 
For, apart from the special question of the genesis of 
respect, the parallelism between the ideas of M. Bo vet and 
those of Durkheim seems to us to be complete. Durkheim 
explains duty under its heteronomous form by means of 
the social constraint of conformist communities, whereas 
he a.ccounts for the development of personal conscience 
by social condensation and differentiation and the result- 
ing individualism. In the same way, Bovet explains ele- 
mentary duties by the respect felt by the little for the big, 
and the progressive autonomy of conscience by the fact of 
the clash of the influences received. Thus the two authors 
use different languages to describe the same mechanisms. 

But if the languages adopted are parallel and not contra- 
dictory to each other, it should be added that psychological 
explanation cannot account for all aspects of moral develop- 
ment without taking into consideration the general shape 
of different societies as a whole. It seems to us that two 
problems in particular must be dealt with in connection 
with M. Bovet's theory the problem of filial respect and 
the problem of the liberation of individual minds. 


We can certainly say, in spite of Durkheim's objec- 
tions, that in a little child's respect for his parents and in 
filial piety in its simplest form we have psychological 
data that are independent of the structure of the sur- 
rounding society. But if the parents are really gods for 
the child, and the rules of primitive morality can all be 
reduced to the parents' commands, one may wonder how 
progress is possible, not so much in the individual in our 
society but in the actual history of moral ideas. In our 
modern communities an individual has only to discover 
other points of view than those to which he is accustomed 
and thus to become conscious of a moral ideal that is 
superior to persons, and he will soon be capable of judging 
his parents, he will cease to deify them and will thus liberate 
his mind and make it accessible to innovations. But in 
primitive gerontocratic communities (and if M. Bovet 
is right the more primitive a community, the more geron- 
tocratic it must be), how will the mind ever come to con- 
ceive of a moral ideal that is superior to accepted custom ? 
Even where the filial feelings are transferred to the col- 
lective religious symbols, the gods can hardly be better 
than men, and mystical rules than the dictates of common 
use. In point of fact, not till the advent of the most 
developed religions have men conferred upon the gods a 
moral purity that was without alloy. However desirous 
elementary religions may often appear of attaining this 
purity, the morals and wishes of primitive deities are 
singularly like the prevailing customs and rules of the 
community in question. How are we to explain the fact 
that humanity ever emerged from this state of affairs ? 

It is at this point that it seems to us all-important to 
consider the general form or shape of societies. In order 
to raise the moral level of the gods and to leave behind 
one the morality of obligatory commands for the sake of 
the morality of an autonomous conscience, a dual process 
is necessary : there must be a spiritualization of filial 
respect and a liberation of individual minds. Such 
processes can be accounted for, according to M. Bovet, 
2 B 


by the fact that the influences brought to bear cut across 
each other ; and the first condition for this to take place 
effectively is that the social environment should be 
sufficiently dense and sufficiently differentiated. As 
Durkheim has shown in a passage that has become classical, 
social condensation alone and the resulting differentiation 
are capable of explaining the liberation of individual 
minds. Thus, and thus only, will the individual be capable 
of judging the commands he has received from earlier 
generations. Then only will filial respect submit to the 
control of reason, and moral consciousness place above 
and beyond persons an ideal of good that transcends all 
duties and all commands. 

But in thus completing M. Bovet's point of view with 
that of Durkheim have we really disposed of all our 
difficulties ? The moment has come for us to return to 
the child and to compare the theories we have been dis- 
cussing with the result of our previous enquiries. We can 
put the matter in a nutshell by saying that M. Bovet's 
doctrine seems to us completely to conform to the facts 
concerning the starting-point of child morality, but 
when it comes to the evolution of conscience in the child, 
the only way to be faithful to the spirit of this doctrine 
is to extend it and to distinguish two types of respect. 

We are faced here with a difficulty that is exactly 
analogous to that which was raised by Durkheim's point 
of view a circumstance sufficient in itself to confirm the 
parallelism between the two points of view. How, we 
may ask, if all his duties come from personalities that 
are superior to him, will the child ever acquire an auto- 
nomous conscience ? Unless we assume something more 
than the morality of pure duty, such a development seems 
to us quite inexplicable. Since tfie content of these 
duties conforms by definition to the rules accepted by 
the parents themselves, it is impossible to see how the 
morality of duty would ever authorize the child to modify 
these rules and to criticize his parents : the formation 
of an inner ideal, that is to say, the morality of good, 


seems to be the only thing that will account for this 
phenomenon. Now, does the clash of influences received 
suffice, together with the intervention of reason, to ex- 
plain the appearance of this ideal ? It would seem that 
it does not. It is easy to see how under the influence of 
contradictions due to commands interfering with each 
other reason will assume the right to define its duties 
more clearly, to generalize their contents, in a word, to 
polish and codify the material of morality. But according 
to M. Bovet's hypotheses, reason can prescribe nothing. 
It speaks in the indicative mood, not in the imperative. 
In short, there is no way out of the heteronomy that 
belongs to the play of commands, even if this play be 
indefinitely complicated ; only by attributing legislative 
power to reason can we account for autonomy. 

But M. Bovet, differing in this from Durkheim, who 
did everything to make his system a self-contained whole, 
has left the road open, and even invites us to extend his 
analyses. Not only has he always drawn a distinction 
between the sense of duty and the feeling for good, with- 
out subsequently trying to identify these two irreducible 
realities, but in addition to this, by representing respect 
to us as a relation between one person and another that 
is capable of various possible combinations, he invites 
us to think of respect as itself becoming differentiated in 
the context of concrete psychological states. 

And this is why, alongside of the primitive respect 
felt by the inferior for the superior, or, as we have called 
it, " unilateral respect ", we have claimed to distinguish 
a " mutual " respect towards which the individual tends 
when he enters into relation with his equals, or when his 
superiors tend to become his equals. The quasi physical 
element of fear which plays a part in unilateral respect then 
gradually begins to disappear in favour of the purely moral 
fear of falling in the esteem of the respected person. The 
need to be respected thus balances that of respecting, 
and the reciprocity resulting from this new relation is 
sufficient to abolish all element of constraint. At the 


same time, the commands vanish and turn into mutual 
agreement, and rules that have been freely consented to 
lose their character of external obligation. Nor is this 
all. For since the rule is now subjected to the laws of 
reciprocity, it is these same rules, rational in their essence, 
that will become the true norms of morality. Hence- 
forward reason will be free to lay down its plan of action 
in so far as it remains rational, that is to say, in so far 
as its inner and outer coherence is safeguarded, i.e. 
in so far as the individual can adopt a perspective such 
that other perspectives will accord with it. Thus out of 
anomy and heteronomy, autonomy emerges victorious. 

But need we make things so complicated, and would 
not M. Bovet's language suffice to describe the same 
facts ? Let us try to translate our observations into terms 
of simple (unilateral) respect and see whether our own 
dualistic terminology is an advantage or a disadvantage. 

Let us imagine the case of two children of the same age 
and both belonging to that stage of codification of rules 
the characteristics of which we observed most clearly 
round about the age of n. It is perfectly true that 
while respecting the personality of the other, each child 
may in turn outstrip his playmate if, indeed, his 
prestige does not eclipse the latter's once and for all. 
Thus there is never complete equality, except in theory 
and as an ideal. We might therefore describe the facts 
of mutual respect as follows : i A gives a command 
to B. 2 B accepts this command because he respects 
A (unilateral respect). 3 But A puts himself mentally 
in B's place (this would be the new fact marking a 
departure from the egocentrism of the initial stages). 
4 A therefore feels bound himself by the command he 
has given B. Moreover, just as one's duties towards 
oneself are the result, thanks to a sort of transference, 
of one's duties towards others, so every kind of mutual 
respect could also be regarded as due to a series of trans- 
ferences of unilateral respect. Respect, limited in the first 
instance to the parents only, would be transferred by the 


child to adults in general, then to older children, and finally 
to his contemporaries, so that the contents of duty would 
be progressively extended to include all individuals, 
including oneself. Thus all forms of respect would, in 
so far as they were sources of moral obligation, derive 
from unilateral respect, and their common character 
would find an explanation in the original object of this 
primitive respect. 

This terminology would certainly have the advantage 
of emphasizing the functional unity of the phenomenon 
of respect. In addition, and for this very reason, it would 
tend to bring closer together, but without confusing them, 
the morality of good and that of duty. Duty would 
occur whenever and wherever there was respect, com- 
mands, and therefore rules. On the other hand, the 
sense of the " good " would be the result of the self- 
same tendency which urges individuals to respect each 
other and to place themselves inwardly in each other's 
minds. We distinguished, in connection with cooperation 
and mutual respect, " constituted " rules resulting from 
reciprocal agreements, and " constitutive " standards, or 
norms serving to define the actual laws of reciprocity. 
In M. Bovet's language constituted rules, at whatever 
level in the development of respect they found them- 
selves, would be nothing but duties, and constitutive 
norms would define the good. 

This translation of our results into the language of 
M. Bovet will be sufficient to show how largely our own 
researches have merely been an extension of his, and 
the apparent divergences between us simply a matter of 
language. But this translation, by the mere fact of being 
possible, brings out the conclusion we particularly wished 
to emphasize, namely, the relative heterogeneity of duty 
and good. For, as M. Bovet has himself constantly 
recognized, there is nothing in the actual form of duty that 
forces its contents to conform with good. Duties are not 
obligatory because of their contents but because of the fact 
that they emanate from respected individuals. There 


may therefore be duties that have nothing to do with 
morality (M. Bovet mentions the obligations of custom) 
and duties that are immoral (in so far as they are con- 
trary to the morality of reciprocity). The individual who 
obeys such commands no doubt experiences a feeling of 
doing what is good in so far as he is respecting someone else 
and submitting to someone else out of respect, but this does 
not establish the value of the commands coming from the 
respected individual. Now, if we have thus to distinguish 
on the one hand a unilateral respect as the source of 
duties whose contents are in principle extraneous to the 
good (though in actual fact they may, of course, coincide), 
and on the other, an ideal of reciprocity which defines 
the good itself, such as it will sooner or later appear to the 
moral consciousness, would there not be a certain advan- 
tage in isolating under the name of mutual (respect) 
the special type of respect whose peculiar characteristic 
it is to constitute the feeling of good ? If the good is 
taken, not as a Platonic ideal, which could be unintelli- 
gible from a psychological standpoint, but as a form of 
equilibrium immanent in the mind, it is desirable, so it 
seems to us, to establish a difference in kind between 
unilateral respect, which leads to the recognition of hetero- 
nomous norms, and mutual respect, which recognizes no 
law but its own mutualness and which thus leads to the 
formation of norms that function within itself. We can 
say, of course, that if A respects B, and vice versa, this 
is because A was first respected by B and then placed 
himself at the same point of view as B. But this is a 
completely new operation in relation to respect pure and 
simple ; for if B simply respects A without the feeling 
being returned, he will regard as duties all the commands 
laid upon him by A, however arbitrary they may be and 
however foreign to the laws of reciprocity. But -as soon 
as A identifies himself morally with B, and thus submits 
his own point of view to the laws of reciprocity, the 
product of this mutual respect is bound to be some- 
thing new, because the norms admitted from now 


onwards will necessarily be contained within this very 

We grant, of course, that in point of fact traces of 
unilateral respect and of inter-individual constraint are 
to be found everywhere. Equality exists in theory only. 1 
It may therefore very well be the case that mutual respect 
is never to be found pure and unadulterated, but is only 
an ideal form of equilibrium towards which unilateral 
respect is guided as the inequalities of age and of social 
authority tend to disappear. From this point of view 
M. Bovet's language is no doubt more accurate than 
ours, but this seems to us to be at the cost of certain 
difficulties on other points. For what, after all, is a 
" transference " ? Can we really say that a feeling re- 
mains identical with itself when its object is changed ? 
Are we not the victims of a genetic illusion when we 
postulate the permanence of a tendency throughout its 
history and transformations ? Would it not be better 
to define psychic realities in terms of the system in which 
they are involved at a given moment of their develop- 
ment ? To see the exact proportion in which the syn- 
chronistic and the diachronistic factors are distributed, 
fine shades must be distinguished and minute differences 
taken into account. According to M. Bovet mutual 
respect, having been derived from unilateral respect, 
remains in a sense identical with it or, at any fate, is still 
based upon it. In our view, mutual respect, being in- 
volved in a different system of equilibria, deserves to be 
distinguished from the unilateral variety. 

The only essential point in this whole debate is the 
following. It matters very little whether mutual respect 
and cooperation are real states or limiting forms of equili- 
brium. The unique contribution of cooperation to the 
development of the moral consciousness is precisely that it 
implies the distinction between what is and what ought to be, 

1 On this point we may refer the reader to the discussion which 
we entered upon with M. Blonde! before the Societ6 fra^aise de 
Philosophic (see Bulletin Soc. franf. Phil., pp. 120-3, 1928). 


between effective obedience and an ideal independent of any 
real command. If unilateral respect and social constraint 
were alone at work, moral good would be given once and 
for all under the imperfect and often grotesque forms 
assigned to it by the duties and regulations of existing 
society. But cooperation and mutual respect, in so far 
as they involve indwelling norms that are never ex- 
hausted by "constituted" rules, play an irreplaceable 
part as catalytic agents and give a definite direction to 
moral evolution. From this point of view, it is unnecessary 
to try and determine up to what point these new realities 
diverge from constraint and unilateral respect in the world 
of fact. The one thing we have to remember is that in 
the psychology of norms the successive stages are not 
everything : the direction itself, the vector, as M. Lalande 
would say, counts for more than anything else. And in 
this matter the terminological differences which separate us 
from M. Bovet in nowise bar the way to our fundamental 
agreement as to methods and results. 

been treating the subject in chronological order, we ought 
to have dealt with Baldwin before discussing M. Bovet's 
theories. But though the theory he puts forward is closely 
analogous to that of Bovet, Baldwin's manner of formu- 
lating the problems is far less precise, and it was therefore 
necessary to study M. Bovet's doctrine first in order to 
elucidate that of his predecessor. Moreover, as Baldwin 
stands almost exactly half-way between Durkheim and 
Bovet, we shall be able to deal with him more briefly than 
if we had had to follow step by step the actual development 
of the various theories relating to child morality. 

Baldwin, it will be remembered, claimed that psycho- 
logical and sociological research must work on parallel 
lines because the individual and the communal minds are 
interdependent. Collective consciousness is nothing more 
than " the generalization " of the contents of individual 
consciousness. But conversely (and this is where Baldwin 


seems to be on the side of the sociologists) there is nothing 
present in individual consciousness which is not the result 
of continuous collective elaboration. 1 

The most familiar, and from the point of view of 
Baldwin's moral psychology, the most fundamental ex- 
ample of this solidarity between the social and the 
individual is that of the consciousness of self. What 
could be more intimate and more strictly *' individual " 
in appearance than the feeling of being oneself and dif- 
ferent from others ? Now, in a famous analysis, Baldwin 
has shown that this feeling is really the result of inter- 
individual actions and of imitation in particular. Far 
from starting from any consciousness of self, the baby is 
ignorant of himself as a subject and locates his subjective 
states on the same plane as physical images : this is the 
" projective " stage. How then does the child ever come 
to discover himself ? As far as his own body is concerned 
it is easy enough to see that he does so thanks to a pro- 
gressive comparison of it with other people's bodies, a 
process that is part and parcel of that of learning to 
imitate. It is because it has a visual perception of another 
person's mouth and imitates the movements of this mouth 
that the baby of 10 to 12 months learns to give its various 
buccal sensations an analogous form ; and so on. In the 
same way, with regard to psychical qualities, it is by 
imitating other people's behaviour that the child will 
discover his own. In this way the individual passes to 
the " subjective " stage in which he is conscious of pos- 
sessing an " I " that is identical with that of others. But 
once his attention has been directed upon himself in this 
way, the child becomes capable of the converse process. 
Having little by little come to assign to himself all the forms 
of conduct he has observed ip others, he learns simul- 
taneously to ascribe to others the feelings and motives of 
which he is conscious in himself. In this way there is 
constituted an " ejective " process which in its alternations 
with the other two constitutes the whole of personal life. 

1 See especially J, M. Baldwin *s Psychology and Sociology. 


For, as the shuttle flies backward and forward between 
ejection and imitation, equilibrium is maintained between 
consciousness of self and awareness of others, just as their 
mutual elaboration had been previously ensured by the 
same process. 

After this it will be easy to understand what is Baldwin's 
idea of moral consciousness. Moral consciousness appears 
when the self is no longer in a state of harmony, when 
there is opposition between the various tendencies that 
constitute it (tendencies which, as we have just seen, are 
themselves of external origin). Whence comes this lack of 
harmony ? From the fact that sooner or later the child is 
compelled to obey the adult, and that in obeying he ex- 
periences something quite new. For obedience is neither 
mere imitation nor ej ection. Obedience creates a new self, a 
fraction of the self that dominates the rest. The truth is that 
in the very act of learning to obey the child builds up for 
himself what Baldwin calls an " ideal self ", a self that is 
submissive to the dictates of adults, a self, that is to say, 
traced on the pattern of their superior self. Obedience is 
thus a sort of transcendent imitation accompanied by 
a sui generis subjectivity and ejection, which in fact 
constitute nothing more nor less than moral conscious- 
ness itself and the evaluation of acts in terms of this 
consciousness. This amounts, therefore, to practically 
the same thing as Bovet's idea, according to which 
commands received create a sense of inner obligation. 
Like Bovet, Baldwin lays stress upon the fact that there 
are no innate duties : every obligation is the result of the 
pressure exercised by the social environment. But, adds 
Baldwin, once the words of command have been received, 
they become so thoroughly incorporated that they create 
a new self. " And the sense of this my self of conformity 
to what he teaches and would have me do this is, once 
for all, my conscience." l This ideal self is thus simply the 
self of the father or of any other " copy for imitation ". 

1 J. M. Baldwin, Social and Ethical Interpretations in Mental Develop- 
ment, p. 51, Macmillan, 1897. 


" It is not I, but I am to become it. Here is my ideal self, 
my final pattern, my ' ought ' set before me." * 

This explains the striking resemblance which exists 
between the contents of individual consciences and the 
exigencies of society. The moral sentiment " is in society 
because it is in all the individuals; but it is in each 
individual because it is already in society. It is one of 
these * arguments in a circle ' with which nature so often 
reasons out the development problem " (p. 299). Such a 
proposition seems either a truism or a superficial account 
of the matter. But if we remember Baldwin's profound 
treatment of the development of the self as bound up 
with life in common, we cannot but recognize the value 
of the above formula. For in the adult's own life the moral 
sentiments remain closely dependent upon the opinions of 
others. " Even the more subtle and intimate judgments 
which we pass upon ourselves are liable to the same 
influence : we judge ourselves in some degree by the meed 
of reproach or commendation which we receive from 
others" (p. 312). 

Here, then, is the refutation of those theories which 
attempted to reduce the facts of inner obligation to the 
facts belonging to the psychology of the individual as 
such, as, for example, the facts of habit or sympathy. 
Duty cannot be reduced to our spontaneous habits, says 
Baldwin ; it is a " habit of violating habit " (p. 55). As 
for sympathy, it has of itself nothing moral in the eyes of 
conscience. To be sensitive alone is not to be good ; 
sympathy must be canalized and steadied. Thanks to 
imitation and to " ejection ", sympathy is natural to the 
self. But before this sympathy can acquire a moral 
character there must be a common law, a system of rules. 

How does the individual attain to the consciousness 
of this rule or common norm which defines the good ? 
The starting-point is that the child should become 
accustomed "to the presence of something in him that 
represents his father, mother, or in general the law-giving 

1 J. M. Baldwin, he. cit. p. 36. 


personality ". Moreover, " obligations instead of diminish- 
ing only increase " with age, because the examples to be 
followed grow more and more in number (p. 300). 

During a second phase, the child applies to everyone 
else the moral laws elaborated in this way. For every 
characteristic acquired by imitation gives rise to an 
" ejection ". " After the child has obeyed and learned by 
obedience he himself sets the law of the house for the 
other members of it. And the law then becomes ' common 
law ' " (p. 54). This common law is absolute for -two 
reasons. On the one hand, thanks to the commands 
received or, as Baldwin puts it, to the " words of command " 
coming from the environment, this law is imperative. On 
the other hand, in applying the law to everyone else the 
child discovers that in actual practice no one, not even 
" the persons who make the law ", yield it strict obedi- 
ence. And this discovery, instead of weakening the 
child's belief in the law, leads him on the contrary to place 
the moral law above particular individuals. 

Thanks to this process the nature of the ideal changes. 
Up till now the moral " good " had been tied to con- 
crete examples. " Law, in the child, is personal in all 
his transition period to a true ethical self ; it is an em- 
bodiment, a self which is essentially ' projective ', which 
he cannot represent nor anticipate in detail " (p. 323). 
But in so far as the child realizes that no one in actual 
fact obeys the law, he learns to place the law above 
individual wills and to conceive of it as an absolute 
transcending the whims or the authority of the parents 

Whence a third and final stage, during which the notion 
of an ideal good comes to give a content to the law itself. 
For up till now the law consisted only in a sort of gener- 
alization or rather sublimation of the orders received. But 
once detached from persons, the " words of command " are 
raised to the rank of something absolute. Henceforward, 
thanks to the work of practical intelligence, the actual 
content of the law is elaborated along autonomous lines. 


" When the child reflects on his social relationships and 
arrives at the beginning of a habit of intelligent submission, 
which he then in turn prescribes to others also, he shows a 
new sort of end not before found in Mm. None of the 
partial thoughts none of his private schemes is now Ms 
end ; no person completely fulfils his new ideal, Ms ideal 
of personality, long or very well " (p. 326). In this way 
we can explain both the possibility of moral progress and 
the inner liberty of conscience. The individual " is now 
launched on a sea of intellectual turmoil and endeavour, 
which by its very restlessness and change, its setting of 
ideals and its violation of them, make social life and 
progress possible " (p. 326). 

Such are the main ideas of Baldwin's moral psychology. 
Is this interpretation of the relation between the conscious- 
ness of duty and the morality of good entirely satis- 
factory ? That is what we have now to examine, and in 
doing so we shall try as far as possible to bring back 
the discussion to the ground covered by the concrete 
observations given in the present volume. 

It should be noted in the first place that without the 
intervention of reason, wMch Baldwin particularly stresses, 
the social processes invoked by our author are not suffi- 
cient to explain the development of the autonomy of 
conscience. It is easy enough to see how the child, 
starting from imitation and ejection, will incorporate in 
his self the " words of command " issuing from Ms social 
environment and finally conceive of the law as superior 
to individuals. But without the intervention which Bald- 
win attributes to practical reason, tMs law will never be 
more than a simple generalization of the commands the child 
has received. How does Baldwin escape this conclusion ? 
By bringing in as a sort of adventitious factor the child's 
actual intelligence, which is supposed to remould the 
contents of the commands and build up a personal ideal. 

But presented in this way the reasons Baldwin invokes 
in his explanation of moral development ran the risk of 
appearing heterogeneous to one another. On one side we 


should have the play of social relations, the source of 
conformity and duty, on the other individual intelligence, 
the source of autonomy and the morality of the good. As a 
matter of fact, every reader of Genetic Logic knows what 
pains Baldwin took to account for the evolution of 
childish reason by means of the social factors spoken of 
just now in connection with morality. Thus logic, like 
duty, he tells us, seems at first to be simply the reflection 
of the pressure exercised by those around the child : this is 
the " syndoxic" phase, corresponding to simple obedience 
in the moral sphere. Then through a play of " ejections " 
and of generalizations characterizing discussion, reflec- 
tion, and discursive logic, the rules of logic end, like 
the rules of morality, by becoming superior to individuals : 
this is the " synnomic " phase, corresponding to the inner 
ideal of the morality of the good. 

But thus in invoking, as Baldwin does, childish intelligence 
to explain the liberation of the mind from imperatives of 
external origin, one cannot escape the following contra- 
diction namely that rational logic is itself derived from 
social processes from which it is supposed to free itself 
in matters of moral psychology. In reality, logic and 
morality are entirely parallel, and if one agrees with Baldwin 
that both develop as a function of a collective elaboration, 
one has no right to appeal to the one in order to explain 
the transformations of the other. Both logic and morality 
begin with conformity and end in autonomy. If we 
wish to account for this evolution, we must do so by 
changes that are inherent in social behaviour itself. 
It is a question, therefore, whether Baldwin's social 
psychology really suffices to make us understand why 
constraint gives way to cooperation, or whether Baldwin, 
like so many others, does not in the last analysis tend to 
give priority to the relation of constraint over against 
the relation of cooperation. 

The two chief difficulties of Baldwin's doctrine on this 
point seem to be the following. Baldwin explains the 
early stages of the child's mentality by imitation and 


pressure of the social environment, but he does not take 
sufficient account of the egocentrism inherent in the 
elementary level of consciousness. In dealing with the 
later stages he therefore fails to make the social factor 
sufficiently interior to consciousness, and to see in the 
laws of reciprocity the rational norm that will confer 
upon the common rules the autonomy necessary to this 

We shall deal quite shortly with the first point, as it 
touches more upon the psychology of intelligence than 
upon the theory of the facts of morality. In a famous 
passage Baldwin has admirably shown how " adualistic " 
consciousness is originally, placing as it does on one and the 
same plane what is internal and external, subjective and 
objective, even psychical and physical. But he has not 
noticed, at least so it seems to us, that to such an adualism 
there necessarily corresponds an illusion of viewpoint such 
that the individual places himself at the centre of every- 
thing and explains all things in terms of this ego- 
centric perspective. In the psychology of the intellect it 
is this egocentrism that seems to us to explain the logic 
and the causality peculiar to the child : his difficulty in 
handling relations and in forming objective causal series, 
etc. From the social and moral point of view, it is this 
egocentrism that explains why, though he is so absorbed 
in others that he conforms to examples and commands 
received from without, the child yet introduces into every 
piece of collective behaviour an irreducible element of 
individual interpretation and unconscious deformation. 
Hence the sui generis attitude found among the smaller 
children with regard both to rules of games and to their 
parents' commands an attitude of respect for the letter 
of the law and of waywardness in its application. 

While, therefore, we are in agreement with Baldwin's 
very profound theory relating to the genesis of conscious- 
ness of self, we wish, nevertheless, to add a remark to 
what he says on the rfile of imitation. One can hardly 
deny, indeed, after reading our author's analyses, that the 


self can only know itself in reference to other selves. But 
imitation will never enable us to perceive in ourselves any- 
thing but what we have in common with others. In order to 
discover oneself as a particular individual, what is needed 
is a continuous comparison, the outcome of opposition, of 
discussion, and of mutual control ; and indeed conscious- 
jiess of the individual self appears far later than conscious- 
ness of the more general features in our psychological 
make-up. This is why a child can remain egocentric for a 
very long time (through lack of consciousness of self), 
while participating on all points in the minds of others. 
It is only by knowing our individual nature with its 
limitations as well as its resources that we grow capable 
of coming out of ourselves and collaborating with other 
individual natures. Consciousness of self is therefore both 
a product and a condition of cooperation. 

This has brought us to the second question that of the 
later stages and the autonomy of conscience. Are the 
logical " synnomic " rules and the corresponding moral rules 
as well fitted as Baldwin thinks them to secure the liberty 
of reason ? If we accept the definitions he gives, they 
undoubtedly are. " At the logical stages of social culture 
the individual sooner or later comes to criticize in a 
measure the social formulae and to reject them in such 
and such of their details by making use of his in- 
dividual judgment. In this way the syndoxic becomes 
personal and synnomic" (Thoughts and Things,). But 
if we consider the way in which Baldwin explains the 
genesis and functioning of this social " synnomic " states 
we feel less certain that the processes he invokes are 
really sufficient to account for the contrast existing 
between the results of cooperation and those of con- 
straint. For, suppose we take a command and impose 
it upon a few individuals who, in their turn, will 
impose it upon others, thanks to the mechanism of 
" ejection ". Even if it irradiates the whole of the social 
group and is elevated to the dignity of absolute law 
superior to the individuals, such a rule can acquire no 


new value from the mere fact of its generality. There 
are in every social group usages accepted by all but which 
are none the less irrational or even immoral. How will 
individual criticism, by means of which Baldwin defines 
the " synnomic ", succeed in really transforming the 
content of the " syndoxic " rules into a content that is 
rational and in conformity with the moral ideal ? Only 
on condition that under the " constituted " rules there 
should be at work a system of " constitutive " norms. 
For a set of customs to acquire a new value in the eyes of 
autonomous conscience, it is not enough that these cus- 
toms should have been ratified after discussion by the 
majority or the sum total of the individuals ; this ratifi- 
cation must result from a genuine agreement founded on 
the laws of reciprocity which constitute reason. Of course, 
when Baldwin speaks of individual criticism, of reflection, 
and of discussion, he certainly means reason itself. But 
it would be an illusion to suppose that the play of imita- 
tions and ejections which develop the self and ensure a 
balance between the individual and the social group is 
sufficient to explain the development of these rational 
norms. These mechanisms account for one fact alone 
(incidentally one of fundamental importance), namely, that 
the individual, in spite of his egocentrism, finds himself 
dominated in his earliest years by a system of rules that 
are external to him and imperative in nature. But to 
pass from this to the rational law internal to the mind 
something more is wanted than a mere ratification on the 
part of individual intelligence ; there must be relations 
of a new type between individuals who meet as equals, 
relations founded on reciprocity, relations that will sup- 
press egocentrism and suggest to the intellectual and 
moral consciousness norms capable of purifying the con- 
tents of the common laws themselves. 

6. CONCLUSIONS. The analysis of the child's moral 
judgments has led us perforce to the discussion of the 
great problem of the relations of social life to the rational 
2 c 


consciousness. The conclusion we came to was that the 
morality prescribed for the individual by society is not 
homogeneous because society itself is not just one thing. 
Society is the sum of social relations, and among these 
relations we can distinguish two extreme types : relations 
of constraint, whose characteristic is to impose upon the 
individual from outside a system of rules with obligatory 
content, and relations of cooperation whose characteristic 
is to create within people's minds the consciousness of 
ideal norms at the back of all rules. Arising from the ties 
of authority and unilateral respect, the relations of con- 
straint therefore characterize most of the features of 
society as it exists, and in particular the relations of the 
child to its adult surrounding. Defined by equality and 
mutual respect, the relations of cooperation, on the 
contrary, constitute an equilibrial limit rather than a 
static system. Constraint, the source of duty and heter- 
onomy, cannot, therefore, be reduced to the good and to 
autonomous rationality, which are the fruits of reciprocity, 
although the actual evolution of the relations of constraint 
tends to bring these nearer to cooperation. 

In spite of our wish to confine the discussion to the 
problems connected with child psychology, the reader will 
not have failed to recognize the affinity of these results 
with those of the historical or logico-sociological analyses 
carried out by M. Brunschvicg and M. Lalande. Le 
Progr&s de la Conscience dans la Philosophie occidentale is 
the widest and the most subtle demonstration of the fact 
that there exists in European thought a law in the 
evolution of moral judgments which is analogous to the 
law of which psychology watches the effects throughout 
the development of the individual. Now to indulge in 
philosophic enquiry is simply to take increasing cog- 
nizance of the currents of thought which enter into and 
sustain the states of society itself. What the philosopher 
does is not so much to create something new as to reflect 
the elaborations of the human mind. It is therefore 
of the utmost significance that the critical analysis of 


history of which M. Brunschvicg has put to fresh use 
should have succeeded in bringing to light in the 
evolution of Western philosophic thought the gradual 
victory of the norms of reciprocity over those of social 

As to M. Lalande, what he says on " la Dissolution " 
as also on the social character of logical norms, has shown 
more than any other work on the subject the duality that 
"lies hidden in the word " social ". There are, M. Lalande 
tells us, two societies : existing or organized society, whose 
constant feature is the constraint which it exercises 
upon individual minds, and there is the ideal or assimi- 
lative society, which is defined by the progressive identifi- 
cation of people's minds with one another. The reader 
will recognize here the same distinction as we have been 
led to observe between the relations of authority and the 
relations of equality. 

Some of M. Lalande's minor contentions would, indeed, 
stand in the way of our complete agreement with his 
ideas taken as a whole. It does not seem to us at all 
certain, for example, that " evolution " in the sense of 
progressive organization is necessarily bound up with a 
society based on constraint. The passage from the homo- 
geneous to the heterogeneous which M. Lalande agrees 
with Spencer in taking as the mark of evolution leads no 
doubt to social differentiation. But this differentiation is 
precisely, as the sociologists have pointed out, the con- 
dition of a break with the conformity due to constraint, 
and consequently the condition of personal liberation. 
Moral equality is not the result of an advance towards 
homogeneity, assuming that agreement can be reached on 
the meaning of this word, but of a mobility which is a 
function of differentiation. The more differentiated the 
society, the better can its members alter their situation in 
accordance with their aptitudes, the greater will be the 
opportunity for intellectual and moral cooperation. We 
cannot, therefore, take the identification of minds, which, 
for M, Lalande, is the supreme norm, to be the same thing 



as cooperation. Without attempting to evaluate this 
" vector ", and limiting ourselves to the mere description 
of psychological facts, what the morality of the good 
seems to us to achieve is reciprocity rather than iden- 
tification. The morality of the autonomous conscience 
does not tend to subject each personality to rules that 
have a common content : it simply obliges individuals 
to " place themselves in reciprocal relationship with each 
other without letting the laws of perspective resultant 
upon this reciprocity destroy their individual points of 

But what do these minor discrepancies matter since it 
.is thanks to M. Lalande's teaching that we are able to 
dissociate what the sociologists have so often tended to 
confuse ? And above all, what do the concepts that are 
used in the interpretation of the facts matter, so long as 
the method employed is the same ? For in the work of 
M. Lalande we have an example of that rare thing 
research on the evolution of norms conducted well within 
the limits of the psycho-sociological method. Without in 
any way neglecting the demands of rationality, this great 
logician has been able to discern in intellectual and moral 
assimilation processes admitting of analysis in terms 
of social psychology while implying by their very 
"direction" the existence of ideal norms immanent in 
the human spirit. 

This concordance of our results with those of historico- 
critical or logico-sociological analysis brings us to a second 
point : the parallelism existing between moral and intel- 
lectual development. Everyone is aware of the kinship 
between logical and ethical norms. Logic is the morality 
of thought just as morality is the logic of action. Nearly 
all contemporary theories agree in recognizing the exist- 
ence of this parallelism from the a priori view which 
regards pure reason as the arbiter both of theoretical 
reflection and daily practice, to the sociological theories of 
knowledge and of ethical values. It is therefore in no way 
surprising that the analysis of child thought should bring 


to the fore certain particular aspects of this general 
phenomenon. 1 

One may say, to begin with, that in a certain sense 
neither logical nor moral norms are innate in the 
individual mind. We can find, no doubt, even before 
language, all the elements of rationality and morality. 
Thus sensori-motor intelligence gives rise to operations of 
assimilation and construction, in which it is not hard to 
see the functional equivalent of the logic of classes and of 
relations. Similarly the child's behaviour towards persons 
shows signs from the first of those sympathetic tendencies 
and affective reactions in which one can easily see the 
raw material of all subsequent moral behaviour. But an 
intelligent act can only be called logical and a good- 
hearted impulse moral from the moment that certain 
norms impress a given structure and rules of equilibrium 
upon this material. Logic is not co-extensive with in- 
telligence, but consists of the sum-total of rules of control 
which intelligence makes use of for its own direction. 
Morality plays a similar part with regard to the affective 
life. Now there is nothing that allows us to affirm the 
existence of such norms in the pre-social behaviour occur- 
ring before the appearance of language. The control 
characteristic of sensori-motof intelligence is of external 
origin : it is things themselves that constrain the organism 
to select which steps it will take ; the initial intellectual 
activity does actively seek for truth. Similarly, it is 
persons external to him who canalize the child's elemen- 
tary feelings, those feelings do not tend to regulate them- 
selves from within. 

This does not mean that everything in the a priori view 
is to be rejected. Of course the a priori never manifests 
itself in the form of ready-made innate mechanisms. The 
a priori is the obligatory element, and the necessary 
connections only impose themselves little by little, as 

1 We have further developed this point at the Ninth International 
Congress of Psychology which met at New Haven (U.S.A.). See Ninth 
International Congress of Psychology, Proceedings and Papers, p. 339 


evolution proceeds. It is at the end of knowledge and not 
in its beginnings that the mind becomes conscious of the 
laws immanent to it. Yet to speak of directed evolu- 
tion and asymptotic advance towards a necessary ideal 
is to recognize the existence of a something which 
acts from the first in the direction of this evolution. 
But under what form does this "something" present 
itself ? Under the form of a structure that straightway 
organizes the contents of consciousness, or under the form 
of a functional law of equilibrium, unconscious as yet be- 
cause the mind has not yet achieved this equilibrium, and 
to be manifested only in and through the multitudinous 
structures that are to appear later ? There seems to us 
to be no doubt about the answer. There is in the very 
functioning of sensori-motoi; operations a search for co- 
herence and organization. Alongside, therefore, of the 
incoherence that characterizes the successive steps taken 
by elementary intelligence we must admit the existence 
of an ideal equilibrium, indefinable as structure but implied 
in the functioning that is at work. Such is the a priori : 
it is neither a principle from which concrete actions can 
be deduced nor a structure of which the mind can become 
conscious as such, but it is a sum-total of functional rela- 
tions implying the distinction between the existing states of 
disequilibrium and an ideal equilibrium yet to be realized. 

How then will the mind extract norms in the true sense 
from this functional equilibrium ? It will form structures 
by means of an adequate conscious realization (prise de 
conscience). To ensure that the functional search for 
organization exhibited by the initial sensori-motor and 
affective activity give rise to rules of organization properly 
so called, it is sufficient that the mind should become con- 
scious of this search and of the laws governing it, thus 
translating into structure what till then had been function 
and nothing more. 

But this coming into consciousness or conscious realiza- 
tion is not a simple operation and is bound up with a 
whole set of psychological conditions. It is here that 


psycho-sociological research "becomes indispensable to the 
theory of norms and that the genetic parallelism existing 
between the formation of the logical and of the moral 
consciousness can be observed. 

In the first place it should be noticed that the individual 
is not capable of achieving this conscious realization by 
himself, and consequently does not straight away succeed 
in establishing norms properly so-called. It is in this sense 
that reason in its double aspect, both logical and moral, 
is a collective product. This does not mean that society 
has conjured up rationality out of the void, nor that there 
does not exist a spirit of humanity that is superior to 
society because dwelling both within the individual and 
the social group. It means that social life is necessary if 
the individual is to become conscious of the functioning of 
his own mind and thus to transform into norms properly 
so called the simple functional equilibria immanent to all 
mental and even all vital activity. 

For the individual, left to himself, remains egocentric. 
By which we mean simply this Just as at first the mind, 
before it can dissociate what belongs to objective laws 
from what is bound up with the sum of subjective con- 
ditions, confuses itself with the universe, so does the 
individual begin by understanding and feeling everything 
through the medium of himself before distinguishing what 
belongs to things and other people from what is the result 
of his own particular intellectual and affective perspective. 
At this stage, therefore, the individual cannot be conscious 
of his own thought, since consciousness of self implies a 
perpetual comparison of the self with other people. Thus 
from the logical point of view egocentrism would seem 
to involve a sort of alogicality, such that sometimes 
affectivity gains the ascendant over objectivity, and 
sometimes the relations arising from personal activity 
prove stronger than the relations that are independent of 
the self. And from the moral point of view, egocentrism 
involves a sort of anomy such that tenderness and dis- 
interestedness can go hand in hand with a naive selfishness, 


and yet the child not feel spontaneously himself to be 
better in one case than the other. Just as the ideas which 
enter his mind appear from the first in the form of beliefs 
and not of hypotheses requiring verification, so do the 
feelings that arise in the child's consciousness appear 
to him from the first as having value and not as having 
to be submitted to some ulterior evaluation. It is only 
through contact with the judgments and evaluations of 
others that this intellectual and affective anomy will 
gradually yield to the pressure of collective logical and 
moral laws. 

In the second place, the relations of constraint and 
unilateral respect which are spontaneously established 
between child and adult contribute to the formation 
of a first type of logical and moral control. But this 
control is insufficient of itself to eliminate childish ego- 
centrism. From the intellectual point of view this respect 
of the child for the adult gives rise to an " annunciatory " 
'conception of truth : the mind stops affirming what it 
likes to affirm and falls in with the opinion of those around 
it. This gives birth to a distinction which is equivalent to 
that of truth and falsehood : some affirmations are recog- 
nized as valid while others are not. But it goes without 
saying that although this distinction marks an important 
advance as compared to the anomy of egocentric thought, 
it is none the less irrational in principle. For if we are to 
speak of truth as rational, it is not sufficient that the 
contents of one's statements should conform with reality : 
reason must have taken active steps to obtain these 
contents and reason must be in a position to control the 
agreement or disagreement of these statements with 
reality. Now, in the case under discussion, reason is still 
very far removed from this autonomy : truth means 
whatever conforms with the spoken word of the adult. 
Whether the child has himself discovered the propositions 
which he asks the adult to sanction with his authority, or 
whether he merely repeats what the adult has said, in both 
cases there is intellectual constraint put upon an inferior 


by a superior, and therefore heteronomy. Thus, far from 
checking childish egocentrism at its source, such a sub- 
mission tends on the contrary partly to consolidate the 
mental habits characteristic of egocentrism. Just as, if left 
to himself, the child believes every idea that enters his head 
instead of regarding it as a hypothesis to be verified, so 
the child who is submissive to the word of his parents 
believes without question everything he is told, instead of 
perceiving the element of uncertainty and search in adult 
thought. The self's good pleasure is simply replaced by 
the good pleasure of a supreme authority. There is 
progress here, no doubt, since such a transference ac- 
customs the mind to look for a common truth, but this 
progress is big with danger if the supreme authority be not 
in its turn criticized in the name of reason. Now, criticism 
is born of discussion, and discussion is only possible 
among equals : cooperation alone will therefore accomplish 
what intellectual constraint failed to bring about. And 
indeed we constantly have occasion throughout our 
schools to notice the combined effects of this constraint 
and of intellectual egocentrism. What is " verbalism ", 
for example, if not the joint result of oral authority and 
the syncretism peculiar to the egocentric language of the 
child ? In short, in order to really socialize the child, co- 
operation is necessary, for it alone will succeed in deliver- 
ing him from the mystical power of the word of the adult. 
An exact counterpart of these findings about intellectual 
constraint is supplied by the observations on the effect of 
moral constraint contained in the present book. Just as 
the child believes in the adult's omniscience so also does 
he unquestioningly believe in the absolute value of the 
imperatives he receives. This result of unilateral respect 
is of great practical value, for it is in this way that there 
is formed an elementary sense of duty and the first 
normative control of which the child is capable. But it 
seemed to us clear thafc this acquisition was not sufficient 
to form true morality. For conduct to be characterized 
as moral there must be something more than an outward 


agreement between its content and that of the commonly 
accepted rules : it is also requisite that the mind should 
tend towards morality as to an autonomous good and 
should itself be capable of appreciating the value of the 
rules that are proposed to it. Now in the case under 
discussion, the good is simply what is in conformity with 
heteronomous commands. And as in the case of intellect- 
ual development, moral constraint has the effect of partly 
consolidating the habits characteristic of egocentrism. 
Even when the child's behaviour is not just a calculated 
attempt to reconcile his individual interest with the letter 
of the law, one can observe (as we had occasion to do 
in the game of marbles) a curious mixture of respect 
for the law and of caprice in its application. The law 
is still external to the mind, which cannot therefore be 
transformed by it. Besides, since he regards the adult as 
the source of the law, the child is only raising up the will 
of the adult to the rank of the supreme good after having 
previously accorded this rank to the various dictates of 
his own desires. An advance, no doubt, but again an 
advance charged with doubtful consequences if coopera- 
tion does not come and establish norms sufficiently inde- 
pendent to subject even the respect due to the adult to 
this inner ideal. And indeed so long as unilateral respect 
is alone at work, we see a " moral realism " developing 
which is the equivalent of " verbal realism ". Resting in 
part on the externality of rules, such a realism is also kept 
going by all the other forms of realism peculiar to the 
egocentric mentality of the child. Only cooperation will 
correct this attitude, thus showing that in the moral 
sphere, as in matters of intelligence, it plays a liberating 
and a constructive r61e. 

Hence a third analogy between moral and intellectual 
evolution : cooperation alone leads to autonomy. With 
regard to logic, cooperation is at first a source of criticism ; 
thanks to the mutual control which it introduces, it 
suppresses both the spontaneous conviction that charac- 
terizes egocentrism and the blind faith in adult authority. 


Thus, discussion gives rise to reflection and objective 
verification. But through, this very fact cooperation be- 
comes the source of constructive values. It leads to the 
recognition of the principles of formal logic in so far as 
these normative laws are necessary to common search for 
truth. It leads, above all, to a conscious realization of the 
logic of relations, since reciprocity on the intellectual plane 
necessarily involves the elaboration of those laws of 
perspective which we find in the operations distinctive of 
systems of relations. 

In the same way, with regard to moral realities, coopera- 
tion is at first the source of criticism and individualism. 
For by comparing his own private motives with the rules 
adopted by each and sundry, the individual is led to 
judge objectively the acts and commands of other people, 
including adults. Whence the decline of unilateral respect 
and the primacy of personal judgment. But in consequence 
of this, cooperation suppresses both egocentrism and 
moral realism, and thus achieves an interiorization of 
rules. A new morality follows upon that of pure duty, 
Heteronomy steps aside to make way for a consciousness 
of good, of which the autonomy results from the acceptance 
of the norms of reciprocity. Obedience withdraws in 
favour of the idea of justice and of mutual service, now 
the source of all the obligations which till then had been 
imposed as incomprehensible commands. In a word, 
cooperation on the moral plane brings about transforma- 
tions exactly parallel to those of which we have just been 
recalling the existence in the intellectual domain. 

Is there any need, by way of conclusion, to point to 
the educational consequences of such observations ? If 
education claims to be the direct application of what 
we know about Child Psychology, it would not be neces- 
sary. It is obvious that our results are as unfavourable 
to the method of authority as to purely individualistic 
methods. It is, as we said in connection with Durkheim, 
absurd and even immoral to wish to impose upon the 
child a fully worked-out system of discipline when the 


social life of children amongst themselves is sufficiently 
developed to give rise to a discipline infinitely nearer to 
that inner submission which is the mark of adult morality. 
It is idle, again, to try and transform the child's mind 
from outside, when his own taste for active research and 
his desire for cooperation suffice to ensure a normal 
intellectual development. The adult must therefore be 
a collaborator and not a master, .from this double point 
of view, moral and rational. But conversely, it would 
be unwise to rely upon biological " nature " alone to 
ensure the dual progress of conscience and intelligence, 
when we realize to what extent all moral as all logical 
norms are the result of cooperation. Let us therefore 
try to create in the school a place where individual experi- 
mentation and reflection carried out in common come 
to each other's aid and balance one another. 

If, then, we had to choose from among the totality of 
existing educational systems those which would best 
correspond with our psychological results, we would 
turn our methods in the direction of what has been called 
" Group Work " and " Self-government/ 1 1 Advocated by 
Dewey, Sanderson, Cousinet, and by most of the pro- 
moters of the " Activity School ", the method of work by 
groups consists in allowing the children to follow their 
pursuits in common, either in organized " teams " or 
simply according to their spontaneous groupings. Tradi- 
tional schools, whose ideal has gradually come to be the 
preparation of pupils for competitive examinations 
rather than for life, have found themselves obliged to 
shut the child up in work that is strictly individual : 
the class listens in common, but the pupils do their home 
work separately. This procedure, which helps more than 
all the family situations put together to reinforce the 
child's spontaneous egocentrism, seems to be jeontrary 
to the most obvious requirements of intellectual and moral 

1 We refer the reader, on this point, to our " Rapport sur les procMes 
de 1'Education morale ", read at the Fifth International Congress on 
Moral Education in Paris, 1930. 


development. This is the state of things which the method 
of work in groups is intended to correct. Cooperation is 
promoted to the rank of a factor essential to intellectual 
progress. It need hardly be said that this innovation 
assumes value only to the extent that the initiative is 
left to the children in the actual conduct of their work. 
Social life is here a complement of individual " activity " 
(in contrast to the passive repetition which characterizes 
the method of teaching by books), and it would have no 
meaning in the school except in relation to the renovation 
of the teaching itself. 

As for self-government, the fine works of F. W. Foer^ter 
(op. cit.) and Ad. Ferri&re l have rendered unnecessary 
the task of reminding our readers of its principles. M. 
Feni&re in particular, has described with great care and 
with that proselytizing fervour which characterizes all 
his educational works the various modes of govern- 
ment of children by themselves. It is hard to read his 
book without being filled both with the hope of seeing 
the experiments he analyses carried out more generally, 
and with the satisfaction at finding in the principles that 
characterize children's republics what we already know, 
thanks to the psycho-sociological study of the moral life. 

As to F. W. Foerster, his moral pedagogy is still in 
our opinion too much tinged with the cult of autho- 
rity or unilateral respect, and, above all, too much 
attached to the idea of expiatory punishment. But this 
makes the preoccupation with autonomy and self-govern- 
ment, which appears in the rest of his work, the more 

But pedagogy is very far from being a mere application 
of psychological knowledge. Apart from the question of 
the aims of education, it is obvious that even with regard 
to technical methods it is for experiment alone and not 
deduction to show us whether methods such as that of work 
in groups and of self-government are of any real value. For, 

1 Ad. Ferri&re, L'Autonomie des colier$, Coll. des Actualit6s pdag. 
Delachaux et Niestl. 


after all, it is one thing to prove that cooperation in the 
play and spontaneous social life of children brings about 
certain moral effects, and another to establish the fact that 
this cooperation can be universally applied as a method 
of education. This last point is one which only experi- 
mental education can settle. Educational experiment, on 
condition that it be scientifically controlled, is certainly 
more instructive for psychology than any amount of 
laboratory experiments, and because of this experimental 
pedagogy might perhaps be incorporated into the body of 
the psycho-sociological disciplines. But the type of ex- 
periment which such research would require can only be 
conducted by teachers or by the combined efforts of 
practical workers and educational psychologists. And it is 
not in our power to deduce the results to which this would 


Activity Schools, 286, 360, 367, 368, 

Aflection, desire for reciprocal, 172, 

the starting-point of the good, 193 
Ambivalence, 190 
Animism, 141, 142, 187, 256, 257 
A priori, 316, 317, 320, 404, 405 ; 

as an ideal equilibrium, 406 
Artificialism. 141, 142, 187, 254, 255, 

256, 257 
Auto-punishment, 258 

Categorical Imperative, 344 

Cheating, at games, 295, 305 ff., 
322 ; at lessons, 285 ff. 

Compassion, 227, 228, 229 

Constraint, adult moral, as a 
socializing factor, 82 ; founded on 
unilateral respect, 79, 95, 103, 315, 
335 349* 3^5> 366, 402 ; cause of 
belief in expiatory punishment, 
227, 230, 297, 340, 341, and in 
immanent justice, 258, 259, 260 ; 
cause of lying, 135, of moral 
realism, 131, 159, 170, 173, 183, 
188, and of revolt, 134, 191 ; 
consolidates egocentrism, 53, 61, 
62, 84-87, 105, 159, 184, 188, 192, 
372, 409, 410 ; inevitable in 
matters of food, sleep, cleanliness, 
etc., 45, 82, 175, 177, 179, 188, 
384 ; parallelism with intellectual 
constraint, 105, 192, 409, 412 ; 
perhaps a secondary growth, 173 ; 
of groups, 96, 102, 187, 202, 203 

Constraint, social, not identical with 
cooperation, 340 ; cause of primi- 
tive ways of thought, 340 ff. ; 
consolidates egocentrism, 350, 402 

Cooperation, as a socializing factor, 
36. 58, 86 ; founded on mutual 
respect, 79, 89, 103, 170, 227, 315, 

335 337 ' 402 ; necessary for 
autonomy, 61, 62, 68, 89, 90, 105, 

335. 337 34, 355* 3^5, 375. *<* 
truthfulness, 161, 164, 168, 171, 
and for personality, 90 ; as a 
rationalizing factor, 85, 103, 184, 
374, 407 ; anti-mystical in ten- 
dency, 54, 56, 76, 95 ; destroys 
egocentrism, 184, 372, 411 ff. ; 
as a state of equilibrium, 26, 84, 
90, 99> 34*> 348, 35> 375, 391, 
402 ; never pure, 79, 84 ; never 
completely realized, 90, 91 ; the 
source of good, 194, 353 ; as a 
method of education, 286 (fpot^ 
note), 412, 413 ; wrongly assimi- 
lated to constraint by Durkheim, 

Democracy, and democratic ideal, 
57, 63, 66, 68, 325, 347, 366 ; in 
schools, 366, 370, 372 

Difficult children, 107, 189 

Discipline, external and internal, 
366, 367, 411 

Education, 40, 41, 80, 165, 192, 
412 ; punishment in, 212, 216, 
231 ; democracy in, 366, 367 ; 
experiment in, 216, 367, 412, 413 

Effort, not identical with obligation, 
352, 369 

Ego, consciousness of, 86, 87, 89, 
90, 249, 393-395* 399, 4, 4* 

Egocentrism, definition of, 26, 249 ,* 
dual character of in practice of 
rules, 16, 18, 19, 24, 26, 27, 31, 
3^. 45. 54* 87, 94, 97 399. 4* ; 
compared to personality, 90 ; 
compared to consciousness of self, 
87. 89, 90, 249, 399, 400. 47 I in 
children's conversations, 32, 76, 

1 This index does not claim to be exhaustive, either in its references or 
in its headings, these being supplementary to those in the Table of Contents. 
It is meant, primarily, to point to the way in which the author has connected 
some of the leading ideas in his book. [Trans.]. 


Egocentrism continued 

85 ; as an intermediate stage, 79, 
85. 94 ; the mark of primitive 
mentality, 26, 77 ; the result of 
constraint, 53, 61, 62, 84-7, 105, 

159, 184, 350, 372, 410, and of 
unilateral respect, 79, 89 ; de- 
stroyed by cooperation, 184, 372, 
410 ff. 

Egocentrism, intellectual, 37, 56, 

160, 161, 162, 399 

Equality, and distributive justice, 
196, 197 ; based on reciprocity, 
198, 273, on cooperation, 274, 
315, and on mutual respect, 279, 


Equilibrium, ideal, consciousness of, 
318 ; a priori as, 316, 405 ; con- 
science as, 341 ; cooperation as, 
26, 84, 90, 99* 34i> 34 8 35<>; 
good as, 355, 390 ; mutual 
respect as, 90, 324, 335, 391 

Equity, as distinguished from 
equality, 216, 271, 282, 283, 310, 
311, 314, 316 

Father, child's ideas about, 47, 48, 


Filial piety, 384, 385 
Finalism, 141, 255, 257 

Generations, pressure of, 68, 100, 

337, 349, 350 
Gerontocracy, 57, 63, 78, 249, 325, 

33^, 385 

Girls, deficient in legal sense, 70, 72, 
75 I greater tolerance of, 73, 
75 ; on " hitting back," 302 

Good, the, xox, 102, 106, 173, 193, 
339, 351-5, 402 ; as an ideal 
equilibrium, 355, 390 ; versus 
duty, 194, 376, 387, 389 ; due to 
reciprocity, 404 ; due to mutual 
respect, 324, 354 

Guilt, sense of, 132, 162, 177 

Hitting back, 297, 301 ff., 322 

Imitation, as socializing factor, 26, 
27 ; in egocentric practice of 
rules, 16, 41, 45, 54, 72, 87, 94, 97 

Inferiority, feelings of, 275 

Initiation, in primitive communities, 

Intelligence, Motor, 21, 22, 79, 81, 

93 ; how related to Rational, 91, 

93, 405 4.06 
Interiorization, of morality, 19, 170, 

182, 193, 332, 399 

Invention, confused with reminis- 
cence, 48, 49, 51, and with divine 
inspiration, 49, 51, 53 

Jealousy, 317 

Judgment, theoretical and practical, 
relations between, 56, 78, 79, 109, 
112-115, 132, 142, 171, 172, 173, 
181, 273 

Legal element, in children's games, 
33, 37 I analogy with formal 
reasoning, 38, 42 ; deficient in 
girls (see Girls) 

Mana, 256 

Mysticism, 49, 78, 95 
Myths, 162, 256 

Oaths, identified with lies, 136 fL, 

Obligation, feeling of, 23, 43, 44 fL, 

81, 100; 193, 354, 373, 375, 377 ; 
founded on respect and com- 
mands, 372, 375, 377 

Onanism, 258 

Parallelism, between moral and 
intellectual facts, 32, 33, 38, 183, 
192, 404, 409, 410 

Parents, early rules imposed by, 2, 
43, 48, 49 ; all rules ascribed to, 
52, 54, 87 ; mistakes of, 130, 134, 
162, 188, 189, 190, 254, 261, 274 ; 
prestige of, 133 

Participation, feeling of, 54, 72, 

Perception, nature of, 186 

Personality, 90 ; distinguished from 
ego, 92, 94 

Physical law, confused with moral, 

82, 83, 185, 187, 188, 189, 255, 
257, 339, 340, 399 ; owing to social 
constraint, 340 ff. 

Primitive mentality, 20, 187, 240, 
339, 349 ; societies, 77, 99, 182, 
240, 249, 340, 365 

Psittacism, 48, no, 172 

Psycho-analysis, 189 

Punishment, arbitrary element in, 
203, 209, 212, 217, 226 ; corporal, 
362, 367, 371 ; as criterion of 
guilt in lying, 165, 168, 187, in 
revenge, 299, in cheating at 
school, 287 ; danger of incon- 
sistent, 134, 266 ; instinctive 
elements in, 227-230, 321 ; as 
marking breach in bond of 
solidarity, 203 fL, as natural 



Punishment continued 

consequence of the act, 205, 206, 
361, 362, 370, 371 ; preventive 
aspect of, 212, 216, 226, 245, 247, 
360 ; vindictive aspect of, 212, 
215 ; whether necessary at all, 
199, 217, 219, 247, 248, 265, 
267 ff., 370 

Punishment, two fundamental types 
of, 198, 207 fi. ; connected with 
two types of morality and respect, 
226 ; law of evolution between, 
224, 225, 230, 261, 262 

Punishment, expiatory, due to 
adult constraint (q.v.), 226, 227, 
230, 273, 297, 321, 324, and adult 
anger, 294, 321 ; related to 
unilateral respect, 322 

Realism, general, 106, 163, 184 ; in 
primitives, 339, 340 ; due to 
unilateral respect, 163, 410 

Realism, moral, definition of, 100 ; 
due to constraint, 179, 188, 193 ; 
due to constraint and unilateral 
respect, 131, 135, 193, 349 ; due 
to constraint and egocentrism, 
I59 170* 182, 192 ; due to verbal 
commands, 17 ; parallelism with 
verbalism, 183, 192, 410 

Realization, conscious (prise de 
conscience), 56, 78, 79, 109, 112, 
173, 181, 406, 411 

Reason and rationality, 95, 102, 103, 
183, 185, 397 ; due to mutual 
respect, 89, 388 ; due to co- 
operation, 103 ; due to re- 
ciprocity, 81, 94, 103, 388, 401, 
402, 407 

Reciprocity, 36, 63-65, 89, 94, 103, 
135, 161, 167, 168, 189, 194, 237, 

294, 387, 390. 399. 4 OI 43 I a 

form of equality, 320 ; based on 

sympathy, 229 
Regelbewusstsein, 23, So 
Regularities of behaviour, in motor 

stage, 20-24, 4 2 * 43* 8* 81, 10* 

Reification, 184, 186 

Religion, 88, 96, 100, 380, 385 ; 

filial origin of, 88, 380 
Respect, sui generis nature of, 379, 


Revenge, 299, 300, 301, 304 

Revolt, 134, 190, 324, 369 

Romancing, 16, 46, 48, 138, 151, 
1 60, 161 

Rewards and punishment, irre- 
levance of to morality, 273, 339 

Sadism, 190 

Schemas, motor, 20, 21, 22, 43, 80, 

8 1 ; ritual, 16, 2026, 42, 43, 44, 

80, 81, 82, 93, 94 
Self, feeling of (see Ego) 
Self-government, 365, 366, 370, 412, 


Sex, 258, 378 

Solidarity between children, 138, 
176, 196, 197, 237, 238, 242, 

. 247, 250, 288, 292, 295, 319 

Spirit of the game, 65, 92 

Symbolism, 21-25, 42 

Sympathy, necessity for in co- 
operation, 133, 134, 1 80, 189, 
344 ; as an instinctive foundation 
for morality, 228-230, 317, 321, 
395, 405 ; lack of in moral 
realism, 182 

Syncretism, 86, 191 

Taboo, 138, 258, 338, 354 

Telling tales, 113, 196, 235 ff., 285, 
288 ff,, 299 ff. 

Time-lag, between thought and 
action, see Judgment and Re- 

Transference, 391 ; of responsibility, 
248, 260 ; of crime, 329 

Verbalism, 106, 191, 192, 410 
Vindictiveness, 227, 228, 229 

" Whys," 180, 187 


Adler, 275 

Antipoff, Mme, 228, 229 

Baechler, Mile, 195, 207 

Baldwin, G. M., 65, 184, 194, 326, 
392 ff- 

Bergson, 77 

Blondel, 85, 391 

Bovet, M., i, 23, 44, 47, 81, 95, 100, 
101, 102, 103, 162, 166, 183, 193, 
194, 261, 279, 318, 321, 326, 351, 

365, 375 ff- 

Brunschwiog, 402, 403 
Biihler, 23 
Biihler, Mme Ch., 82 

Claparede, 56, 173, 184, 191, 355, 


Comte, Auguste, 69, 336 
Cousinet, 9, 412 

Decroly, 191 e 

de Maistre, J., 208 

Descartes, 91 

Descoeudres, Mile, in, 115, 144 

D'Essertier, 100 

Dewey, 355, 367, 412 

Durkheim, i, 13, 23, 49, 81, 95, 96, 
97, 99, 100, 102, 103, no, 183, 
198, 202, 205, 295, 325, 326 &,, 
341 ff., 355 ff., 375 ff., 411 

Fauconnet, 232, 248, 326 ff. 
Feldweg, Mile, 194, 244 
Fernald, 115 
Fernandez, Ramon, 90 
Ferriere, 365, 413 
Foerster, F. W., 366, 413 
Freud, 21, 177 

Gosse, Edmund, 189 
Guyau, 208, 339 

Hall, Stanley, 355 
Ihering, 330 

James, W., 259 
Janet, 161, 166 

Kant, i, 95, 101, no, 340, 357, 374, 

375. 379 
Knapp, 209 

Lacombe, R., 100 
Lalande, 65, 402, 403, 404 
Lambercier, i 
Levy-Bruhl, 260 
Luquet, 185 

Maison des Petits, 226 
Martinez, i 
Maso, N., 104 
Monnier, P., i 
Montaigne, 360, 367 

Pareto, 17, 68, no, 115 

Parodi, 346, 374 

Piaget, J., 37, 41, 49, 5<5, 65, 87, 

116, 141, 176, 186, 187, 259, 404, 

Piaget, Jacqueline, 19, 20, 22, 25, 

71, 73, 175-180 
Piaget, Mme J. V., i 

Rambert, Mile, 195, 251, 263, 284, 

297, 301* 3 8 
Rauh, F., 340, 347, 
Reynier, Mme, 86 
Rousseau, 205 

Sanderson, 367, 412 

Saussure, 21, 383 

Socrates, 346 

Spencer, Herbert, 205, 361, 403 

Stern, 27, 135, 160, 174 

Tolstoi, 368 

Wells, H. G., 367 
Westermark, 328 
Weissfeller, Mme, 180